This manuscript has been reproduced from the microfilm master. UM I films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type o f computer printer.
T he quality o f this reproduction is dependent upon th e q u ality o f th e copy suhm itted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send UM I a complete manuscript and there are missing pages, these will be noted. A lso, if unauthorized copyright material had to be removed, a note will indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand com er and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back o f the book.
Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6” x 9” black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact U M I directly to order.
UMI
A Bell & Howell Information Company 300 North Zeeb Road, Ann Arbor MI 48I06-I346 USA
C H A R A C T E R IZ A T IO N , R E G U L A T IO N AN D E X P R E S SIO N IN TW O SA L M O N ID S
Kristian Robert von Schalbiirg B. Sc., U niversity o f V ictoria, 1991
A Dissertation Submitted in Partial Fulfillment of the Requirements for the Degree of
DOCTOR O F PHILOSOPHY in the Department
o f Biology
W e accept this dissertation as conform ing to the required standard
Dr.N. Sfierwood (Dept, o f Biology)
Dr. F. Choy (Dept, o f Biology)
>r. W. Hinlz~tBept. o f Biology)
Dr. F ^ ano (Dept, of Biochemistry and M icrobiology)
Dr. R. Devlin (External examiner)
© Kristian Robert von Schalburg, 1998 University of Victoria
All rights reserved. This dissertation may not be reproduced in whole or in part, by photocopying or other means, w ithout pennission from the author.
There are currently thirteen members of the gonadotropin-releasing hormone (GnRH) family. The G nRH members that activate the synthesis and release o f the pituitary gonadotropins are the best understood. These members stand central to the development and maintenance o f reproductive function. The roles o f G nR H that act in the brain and not in the pituitary, or that are expressed in extraneural tissues, are not well characterized.
M y goal was to determine whether 1) the regulatory regions and organization o f the GnRH gene is conserved between mammals and fish, 2) G nR H is expressed in tissues other than the brain o f salmonids and 3) the processes that regulate the expression o f GnRH are conserved between two salmonid species with different habitats and reproductive patterns (sockeye salm on, Oncorynchus nerka; rainbow trout, Oncorynchus mykiss).
To determine w hether the regulatory regions and organization o f the GnRH gene were conserved across the species, I isolated and characterized salm on (s)GnRH gene! from rainbow trout and sGnRH geneZ from sockeye salmon. In salm on, which are tetraploid, each duplicated sGnRH gene encoded a different mRNA (m R N A l or mRNA2), but the identical sGnRH peptide. A Southern blot analysis revealed that other related forms o f GnRH exist in the sockeye salmon genome. Also, I determ ined from RT-PCR analysis that GnRH was not expressed in the heart, liver, gut, adrenal, spleen o r retina, but was
expressed in the gonads o f sockeye salmon.
To understand the function o f GnRH in the gonads, it was necessary to learn when GnRH was expressed during development and throughout the reproductive cycle. Studies using RT-PCR analysis and primer extension analysis dem onstrated that the reproductive tissues o f salmonids use an upstream prom oter to regulate G nR H expression. Intron 1 may be retained, resulting in niRNAs containing 5'-untranslated regions longer than their brain counterparts. These sGnRH transcripts lire initiated by a TA TA -less prom oter region from a start site at 315 basepairs upstream from that utilized in the sockeye salmon brain.
Using the same techniques, differences in the expression o f G nRH in em bryonic tissue and gonads o f sockeye salmon and rainbow trout were noted over the first two years of their lives. First, the upstream prom oter is transiently used for expression o f GnRH as early as 14 days after fertilization in rainbow trout and 30 days after fertilization in sockeye salmon. Second, in sockeye salmon ovary and testis, GnRH was expressed in O ctober o f the first year and then only during M ay and June o f the second year in precocious tissue. F or rainbow trout, G nRH was expressed in the first year from M ay to O ctober and in the second year only in Decem ber. Precociously mature ovary and testis expressed GnRH from June to O ctober in the second year.
It was also im portant to determine whether the GnRH mRNA expressed in the developing ovary and testis was translated into protein. H igh pressure liquid
chromatography and radioim m unoassays were used to dem onstrate the presence o f at least three forms o f GnRH in precociously mature ovaries and testes during the second year.
The expression o f sGnRH mRNA2 in the salmonid ovary and testis utilizes an
alternative promoter. The resulting mRNAs have long 5'-untranslated regions that may be im portant in post-transcriptional control. Expression of G nR H in the brain is constant, but is interm ittant in the salm onid gonad. GnRH mRNA is expressed in undifferentiated gonadal tissue in the first year and briefly in differentiated, but im m ature gonads.
However, in precociously mature ovary and testis in the second year G nR H is transcribed and translated at the stage that precedes ovulation and spawning. Differences in pattern and longer duration of G nR H expression are shown in the ovary and testis o f rainbow trout in com parison to sockeye salmon. This might indicate that G nRH is im portant in the regeneration o f new sets o f germ cells in the iteroparous rainbow trout, but not in the semelparous sockeye salmon.
A comparison o f the genes that encode sGnRH m RNAl and m R N A 2 reveals
significant sequence divergence in their 5’-flanking regions following tetraploidization. A large portion o f the sockeye salmon geneZ is missing in comparison to the Atlantic salmon
gene2. However, the salmonid genes all share strong sequence identity in the proximal- prom oter region. Although large segments of sequence identity do not exist in the regulatory regions o f the GnRH-encoding genes o f m am m als and salmonids, some similarities exist in the positions of potential POU-homeodomain regulator and estrogen response elem ent motifs. This suggests that some regulatory control for expression of GnRH in both the brain and gonads may be conserved.
Examiners:
Dr. N. Sherwood (Dept, o f Biology)
Dr. F. Choy (Dept, o f Biology)
Dr. W. Hintz (Dept, o f Biology)
Dr. F. Nano (Dept, o f Biochemistry and M icrobiology)
A B S T R A C T ...ii T A B L E O F C O N T E N T S ... v L I S T O F F I G U R E S ...viii L I S T O F T A B L E S ...xii L IS T O F A B B R E V IA T IO N S ... xii A C K N O W L E D G E M E N T S ... xiv C H A P T E R 1: General In troduction... 1 GnRH structure and fo rm ...2
M ultiple GnRH forms in single sp ec ie s... 7
G nRH cD NA tran scrip ts... 13
G n R H g e n e s in v e rte b ra te s ...19
GnRH fu n c tio n ...23
A. Release o f gonadotropin... 24
B. Release o f growth hormone and p ro lac tin ...25
C. Putative effects on behaviour... 25
D. Local action in reproductive o rg a n s ...26
E. Functions in ancestral anim als... 26
Purpose o f th e sis... 27
C H A P T E R 2: C haracterization o f the Pacific salmon gonadotropin-releasing
horm one gene, co p y num ber and transcription start site... 45
S um m ary... 46
Introduction... 47
Materials and M ethods... 48
R esults... 52
D i s c u s s i o n ... 62
Literature C ite d ...69
C H A P T E R 3: D ifferences in the regulation and expression o f the gonadotropin-releasing hormone gene in the brain and gonads in rainbow trout...73
S um m ary... 74
Introduction...75
Materials and M ethods... 77
R esults... 80
D i s c u s s i o n ... 101
Literature C ite d ...109
C H A P T E R 4: R egulation and expression o f gonadotropin-releasing hormone in embryonic and gonadal tissue o f sockeye salm on...114
S um m ary... 115
Introduction... 116
Materials and M ethods... 117
R esults... 119
D is c u s s io n ... 139
C H A P T E R 5: Characterization o f im m unoreactive G nRH peptide
in the brain and gonads o f rainbow tro u t... 147
S um m ary...148
Introduction...149
Materials and M ethods...150
R esults... 154
D i s c u s s i o n ... 166
Literature C ite d ...171
CH APTER 6: General C o n clu sio n s...175
C onclusions...176
A. Development o f the brain and g o n a d ... 176
B. Effect of GnRH on gonad p h y sio lo g y ... 177
C. Differences in G nR H p ro m o ters...179
D. Purpose o f an alternative prom oter in transcription o f G nR H in gonads 188 Future studies...189
A. Localization o f distinct sG nR H transcripts in specific ce lls...189
B. Interaction o f GnRH(s) with its receptor in reproductive tis s u e s ... 190
C. Determ ination o f function o f GnRH in reproductive tissues... 191
D. Connection o f GnRH effects with arachidonic acid metabolism in g o n a d . . . . 192
E. Association o f GnRH expression with growth in reproductive tissues 193 Literature cited ... 194
LIST O F FIGURES
C H A P T E R 1
Figure 1-1: Com parison of thirteen GnRH p e p tid e s ...4
F igure 1-2: D iagram o f the m ajor groups o f living teleosts... 10
Figure 1-3: Diagram showing location o f different G nR H forms in brain se c tio n s 12
Figure 1-4: A com parison of known preproG nR H s at the am ino acid level... 15/16
Figure 1-5; Comparison o f the genes for five distinct forms of G nR H ... 21
C H A P T E R 2
Figure 2-1 : The com plete nucleotide sequence o f the sGnRH gene2 isolated
from sockeye salm on...54 Figure 2-2: Nucleotide sequences o f the sG nRH -encoding cDNA
isolated from sockeye salmon b ra in ...56 Figure 2-3: Determ ination o f the transcription start site o f the sGnRH gene2 by
prim er extension in sockeye salmon b ra in ...59 F igure 2-4: A nucleotide sequence com parison of the 5'-flanking regions of
Pacific (P) (sockeye) and Atlantic (A) salmon sGnRH gene2... 61 Figure 2-5: Southern blot analysis to detennine GnRH gene copy num ber in
C H A P T E R 3
Figure 3-1: Nucleotide and amino acid sequence com parison o f sGnRH
cDNA 1 and cDNA2 isolated from rainbow tr o u t...82 Figure 3-2: Schem atic showing positions of primers used in RT-PCR
and four different sGnRH cDNA2 transcripts isolated from g o n a d s... 85 Figure 3-3: R T-PCR analysis o f RNA extracted from tw o-year-old rainbow trout
ovary, testis and brain... 87 Figure 3-4: N ucleotide sequence o f the full-length sGnRH cD N A 2 transcript
iso lated from rainbow trout ovary and te stis...90 Figure 3-5: D eterm ination o f the upstream transcription start site used by sGnRH
cD N A 2 transcripts in trout ovary and te stis... 92 Figure 3-6: N ucleotide sequence o f the 5'-flanking region o f the rainbow trout
sG nR H g e n e ! ... 95 Figure 3-7: A schem atic comparing the 5'-flanking regions o f the known
salm onid sGnRH-encoding g e n e s... 97 Figure 3-8: G nRH expression in ovary and testis o f rainbow trout in
first and second year o f their liv es...100 Figure 3-9: Exam ination of GnRH expression and use o f the upstream prom oter
in two-year-old rainbow trout ovary (O), testis (T) and brain (B) with prim er sets A/C and B /C ... 103
Figure 4-1: Tissue-and stage-specific study o f various sockeye salmon tissues
examined by R T -P C R ... 121 Figure 4-2: Analysis o f the upstream transcription start site o f the
salm onid G nR H g e n e ...124 Figure 4-3: GnRH expression using RT-PCR in reproductive tissues of
sockeye salm on in first and second years o f their liv e s ...126 Figure 4-4: RT-PCR analysis o f RNA extracted from two-year-old salmonid
ovary (O), testis (T) and brain (B )... 128 Figure 4-5: Nucleotide sequences o f the full-length sGnRH cD N A 2 isolated from
ovary and testis o f sockeye salm on... 130 Figure 4-6: Nucleotide and am ino acid sequence comparison o f the
sGnRH cD N A l and cDNA2 transcripts isolated
from sockeye salmon and rainbow tro u t... 134 Figure 4-7: RT-PCR analysis o f RNA extracted from sockeye salmon embryos
during early stages of developm ent... 136 Figure 4-8: RT-PCR analysis o f RNA extracted from rainbow trout em bryos
C H A P T E R 5
Figure 5-1: Immunoreactive GnRH in 17-month-olcl immature and
precocious ovary and testis e x tra c ts... 156 Figure 5-2: Immunoreactive GnRH in 1 8 -to 20-month-old precocious ovary
and testis e x tra c ts...158 Figure 5-3: Immunoreactive G nRH in 21-m onth-old maturing ovary and testis
extracts after H PLC elution... 161 Figure 5-4: Im m unoreactive GnRH in 17-month-old brain extracts showing
H PLC positions and am ounts o f im m unoreactivity...163 Figure 5-5: Imm unoreactive G nRH in 21 -m onth-old brain extracts showing
H PLC positions and am ounts o f im m unoreactivity...165
C H A P T E R 6
Figure 6-1: Nucleotide sequence comparison o f the mammalian and
salm o n id G nR H p ro x im al p ro m o te rs... 182 Figure 6-2: Comparison o f mamm alian and salmonid GnRH consensus sequences
for the proximal prom oter in Fig. 6 - 1 ... 184 Figure 6-3: A schematic com paring positions o f potential recognition motifs
in the 5'-flanking regions o f mammalian and
LIST O F TABLES
C H A P T E R 5
T A B LE 5-1: Details of fish and am ount of GnRH peptide per o rg a n . 152
LIST O F ABBREVIATIONS
G o n a d o tr o p in -re lea sin g h o rm on e (G n R H ) fam ily:
cG nRH-I: chicken GnRH-I
cG nRH-II: chicken GnRH-II
cfGnRH: catfish GnRH
dfGnRH: dogfish GnRH
gpGnRH: guinea pig GnRH
hGnRH: herring GnRH
(hyp9)mGnRH: hydroxyproline-9 mammalian G nRH
IGnRH-I: lam prey GnRH-I
IGnRH-III: lamprey G nR H -llI
mGnRH: mammalian GnRH
sGnRH: salmon GnRH
sbGnRH: seabream GnRH
tGnRH-I: tunicate GnRH-I
T ech n iq u es and ch em ica ls used in the isolation o f G nR H :
DTT: dithiothreitol
EDTA; disodium ethylenediamine tetraacetlc acid
HPLC: high pressure liquid chromatography
KCI: potassium chloride
MgCl2: magnesium chloride
PCR: polymerase chain reaction
RIA: radioimmunoassay
RT-PCR: reverse transcriptase-PCR
Tris-HCl: tris(hydroxymethyl)aminomethane O th er related term s:
aa: amino acids
bps: basepairs
CNS: central nervous system
cDNA: complementary DNA
DAF: days after fertilization
DNA: deoxyribonucleic acid
dN TPs: deoxyribonucleo triphosphates
ERE: estrogen response element
FSH , GTH-1: follicle stimulating hormone, gonadotropin-I
GAP: GnRH-associated peptide
irGnRH: immunoreactive GnRH
IGF-I: insulin growth factor-1
LH, GTH-II: luteinizing hormone, gonadotropin-II
mRNA: messenger ribonucleic acid
nts: nucleotides
ACKNOWLEDGEMENTS
This thesis has been like creating an artwork, not really knowing the final outcome, but beginning with one brushstroke to ultimately culminate in a beautiful painting. Actually, it was really more like a wonderful symphony because it required the playing o f so many different instruments to come to fruition. First and foremost, I would like to thank Bill H arrow er, Jack Nickolichuk, Kevin Nickolichuk and Jim Powell. W ithout their help in providing sockeye salmon or rainbow trout this project would never have been completed. Also, three important people need special mention. Sandra Krueckl and D ave Lescheid for being faithful and true friends in every measure o f the word and Carol W arby whose pleasant personality and excellent technical assistance made the HPLC/RIA project a real joy.
I also feel truly blessed to have had the opportunity to work in a lab in which all o f my colleagues were infonriative, good-natured and generous. You know w ho you are, in no particular order— Yogi Carolsfeld, Kathryn Clark, Imogen Coe, Kevin Cum mings, Pam Gillis, Erica Fradinger, Sarah Leary, Nola Erhardt, John M cRory, Dan O 'N eill, Dave Parker, Mary Power, M ike Roch, Emily Standen and Annika Stein. I also would like to thank Brian Antonson and Steeve Pepin for their superior com puter skills and friendship. Special thanks go to H eather Down and Tom Gore for their help in figure and slide presentation and to all the folks in BioStores, particularly Sarwan Dillon, John Morrison and Melinda Powell, who went that extra mile with a smile. 1 also am particularly appreciative o f my friends and family for their understanding and support.
Last, but not least, 1 would like to thank my supervisor, Nancy Sherw ood, for all of her support and encouragement during the last seven years. I greatly adm ire her energy, focus and devotion to her teaching, graduate students and to scientific research. Her ability to draw out the important, the positive and the novel in research and for its dissemination is outstanding.
There will always be a special place in my heart for each one o f you and I pray that the T ruth you seek shall be found.
General Introduction
Parts o f this chapter have been modified and published as:
Sherwood, N.M ., von Schalburg, K. and Lescheid, D.W. (1997). O rigin and evolution o f G nR H in vertebrates and invertebrates. In: GnRH Neurons: G ene to B ehaviour (I.S. P arhar and Y. Sakum a, eds.). B rain Shuppan Publishers, Tokyo, pp. 3-25.
Gonadotropin-releasing horm one (GnRH) is a neuropeptide that is central to the developm ent and m aintenance o f reproductive function. Synthesized and secreted from nerve cells, GnRH acts on a subset o f pituitary cells known as gonadotropes. Upon binding to gonadotrope receptors, G nRH activates the synthesis and release o f follicle stim ulating hormone (FSH) and luteinizing horm one (LH) in tetrapods. In fish, the com parable pituitary hormones are gonadotropin-I (GTH-I) and gonadotropin-II (GTH- II), which are biologically sim ilar to the tetrapod FSH and LH, respectively (Swanson et
al., 1991). These hormones, in turn, stim ulate the production o f gonadal androgens and
estrogens, which feedback to the brain and pituitary to com plete the hypothalam ic- pituitary-gonadal system and regulate sexual behaviour and reproductive cycles.
Currently eleven distinct gonadotropin-releasing hormone (GnRH) structures have been reported for vertebrates and two for invertebrates (Fig. 1-1). T he first GnRH peptide structure was determ ined for m am m als, specifically for pig (M atsuo e t at., 1971) and sheep (Burgus et al., 1972). Since its first characterization, G nR H has now been isolated and sequenced from representatives o f all seven classes of vertebrates. The structural pattern o f the known G nR H peptides is clear: the length is ten am ino acids and four o f the am ino acids are identical as shown in Fig. 1-1. A m ino acid 1 (the amino terminal), am ino acids 9 and 10 (carboxy term inal) and am ino acid 4 are conserved, even in the two new GnRH fam ily mem bers identified in a protochordate, the tunicate. Am ino acids 1, 2 or 3 are im portant for receptor activation and residues 5 to 8, the most variable region, are im portant for specific binding within the particular species (Sealfon
MAMMAL pG L U H I S TRP S E R TYR GLY LEU ARG PRO GLY
GUINEA PIG pG L U T Y R TRP S E R TYR GLY VAL ARG PRO GLY
PRO GLY
CHICKEN-I pG L U H I S T R P S E R TYR GLY LEU
SEABREA M pG L U H I S TRP S E R TYR GLY LEU
HERRING pG L U H I S TR P S E R
CATTISH pG L U H I S TR P S E R H I S GLY LEU
H I S GLY LEU GLN S E R S E R A S N PRO GLY RO GLY PRO GLY
SALMON pGLU H I S TRP SER TYR GLY T R P L E U PRO GLY
DOGFISH pGLU H I S TRP S E R H I S GLY T R P L E U PRO GLY
CHICKEN-II pGLU H I S TRP SE R H I S GLY T R P TY R PRO GLY
LAMPREY-III pG LU H I S TRP S E R H I S A S P T R P LY S PRO GLY
LAMPREY-I pGLU H I S TYR S E R L E U GLU T R P LY S PRO GLY
TUNICATE-I pG LU H I S TRP S E R A S P TY R P H E LY S PRO GLY
40,(XX)) o f vertebrate species. T he reason is that m any species have the identical forms o f GnRH as show n by high pressure liquid chrom atography (H PLC ), radioim m unoassay (RIA) and in som e cases by chemical sequencing (see S herw ood et al., 1997). For exam ple, the m am m alian GnRH (mGnRH) m olecule has been isolated and sequenced as a peptide from the hum an placenta (Tan and Rousseau, 1982), and from the brain o f pig (M atsuo et al., 1971), sheep (Burgus e ta l., 1972), frog (C onlon et al., 1993) and sturgeon (Lescheid e t al., 1995). The chicken GnRH-1 (cGnRH-1) peptide was
sequenced from the brain extracts o f chicken (King and M illar, 1982a and b, M iyamoto
e t al., 1982 and 1983) an d alligator (Lovejoy et al., 1991a).
The chicken GnRH-11 (cGnRH-11) form is present in representatives o f the
cartilaginous fish, bony fish, am phibians, reptiles and birds. F o r exam ple, cGnRH-11 has been isolated and sequenced from the brain in a num ber o f species: chicken (M iyamoto
et al., 1984), alligator (Lovejoy et al., 1991a), frog (C onlon et al., 1993), sea bream
(Powell et al., 1994), tilapia (Parhar, 1997), pacu (Pow ell et al., 1997), catfish (Ngam vongchon et al., 1992 and Bogerd et al., 1992), dogfish (Lovejoy et al., 1992) and ratfish (Lovejoy e ta l., 1991b). Although cGnRH-11 has not been isolated and sequenced as a peptide from any mammal, there is H PL C and RIA data suggesting that this form o f G nRH is present in mamm als including m onotrem es (K ing et al., 1994), m arsupials (King e t al., 1989, 1990, 1994), prim itive placental m am m als (D ellovade et
al., 1993) and prim ates (Lescheid e ta l., 1997). H ow ever, cGnRH-11 is not present in all
vertebrates as it has not been detected in jaw less fish such as lam prey (Sherw ood et al., 1986 and Sow er et al., 1993). T o date the evidence shows that cGnRH-11 has been more strongly conserved than any other GnRH peptide in vertebrates.
O ther G nR H peptides that appear to be confined to various fishes include salmon GnRH (sGnRH), isolated and sequenced from salmon (Sherw ood e ta l., 1983), herring (Carolsfeld, personal com m unication), pacu (Pow ell et al., 1997) and tilapia brain
al., 1994), cichlid (Powell ct al., 1995), pacu (Powell et al., 1997) and tilapia (Parhar,
1997). Catfish G nRH (cfG nRH ) is identified in catfish only (N gam vongchon e ta l., 1992 and Bogerd e ta l., 1992). D ogfish GnRH (dfGnRH) is from dogfish only (Lovejoy
e ta l., 1992). Lam prey GnRH-I (IGnRH-I) and GnRH-III (lam prey G nR H -II has not
been sequenced) are identified only from lam prey (Sherwood et al., 1986 and Sow er et
al., 1993).
A m odified form o f G nR H was reported that did not involve a novel structure, but rather a post-translational change o f one o f the existing forms o f G nR H . This altered form o f G nR H involves m G nR H with Pro^ replaced by hydroxyproline (H yp^). This form, (H yp^)m GnRH, was detected by HPLC and RIA in hum an, rodent, ovine and am phibian hypothalam ic extracts (Gautron et al., 1991).
The m ost recently characterized G nR H structures are the ones isolated from an invertebrate, the sea squirt (Protochordata; Tunicata; Chelyosoma productum ) as shown in Fig. 1-1 (Powell e t al., 1996). Both tunicate GnRH-I (tGnRH-I) and tunicate GnRH- II (tGnRH-Il) are 60% identical to m am m alian G nRH and therefore are m em bers o f the GnRH fam ily based on structural identity. tGnRH-I more closely resem bles IGnRH-I and IGnRH-III because o f the presence o f Lys^, which tGnRH-II lacks. The unusual aspect, however, is that tG nR H -II w as isolated as a dimer. The two chains o f the dim er are identical and linked by a cystine in position six. Tunicate G nR H -I has a structure that is distinct from the m onom er o f tGnRH-II and it is therefore predicted that the two peptides will have different functions.
The evidence that G nRH evolution predates chordate evolution is provided by the tunicate GnRH peptides and the presence o f an im munoreactive G nR H -like m olecule in the CNS neurons o f the gastropod m ollusc, Hellsoma trivolvis (G oldberg e ta l., 1993).
Initial studies done in the 1970s indicated that only a single form o f GnRH, nam ed luteinizing horm one-releasing horm one (LH RH) or m am m alian GnRH, was present in individual vertebrate brains. H ow ever, reports o f GnRH structures distinct from
m G nRH appeared in the early 1980s. T he first indication that a species could have m ore than one form o f G nRH came from studies showing that chicken (M iyam oto et al., 1983 and 1984) and salmon (Sherw ood e t al., 1983) each have two forms o f G nRH. Chicken GnRH-1 an d GnRH-11 were sequenced from peptides isolated from the chicken brain (M iyam oto e ta l., 1983 and 1984); salm on GnRH (Sherwood e ta l., 1983) and a peptide identical to chicken GnRH-11 (Powell, personal com m unication) were sequenced from the chum salmon brain.
These three GnRH structures (cG nRH -I, cG nR H -lI and sGnRH), which w ere distinct from m G nR H , established the two im portant principles that G nR H is a fam ily o f
peptides and a single species can have m ore than one form o f GnRH. H ow ever, it was not know n whether individuals within the species had more than one form o f G nR H in the brain because pooled brains were used for the purification procedures. A t present, the best proof that single species have at least two forms o f G nR H com es from studies in which two G nRH peptides have been isolated from a species and sequenced. These studies include:
lam prey {Pelromyzon m arinus) IGnRH-I and IGnRH-III (Sherwood et al., 1986, Sow er et a /.,1993) dfG nRH and cG nRH-II
dogfish {Squalus acanthias) salm on (Oncorhynchwi keia) catfish (Claria.^, two species)
frog (R ana ridibunda) alligator (A. mississippicnsis) chicken (G allus domesticus)
(Lovejoy et al., 1992) sGnRH and cGnRH-II (Sherwood et at., 1983; pers. com m un.) cfG nRH and cGnRH-II
m G nRH and cGnRH-II cG nRH-I and cGnRH-II cG nRH -I and cGnRH-II
(Bogerd e /a / ., 1992) (Ngam vongchon et al., 1992) (Conlon et ai., 1993)
(Lovejoy et al., 1991a)
the brain tissue o f single fish species in the orders Characiform es and Perciform es (Groups 5 and 10 in Fig. 1-2). For tilipia (Perciform es: O reochrom is niloticus), three G nR H form s have been sequenced as peptides: sea bream G nRH (sbGnRH), sG nRH and cG nR H -II (W eber et al., 1997). For gilthead sea bream (Perciform es: Sparus aurata), these three form s o f G nR H are also present o f which sbG nR H and cGnRH-11 were sequenced as peptides and sGnRH identified by HPLC-RIA (Powell e ta l., 1994). All three G nR H structures w ere subsequently confirm ed by com parison to cD NA sequences (G othilf e ta l., 1995 and 1996). For cichlids (Perciform es: H aplochrom is burtoni), the sam e three G nR H form s are present as shown by H PLC -R IA data (Powell et al., 1995). T o confirm these identities, the sbGnRH peptide was sequenced (Powell et al., 1995) and all three cD N A sequences were obtained (see W hite, S.A. et al., 1995). In another order o f fish, C haraciform es, the same three G nR H peptides have been isolated and sequenced for the fish known as pacu (P iaractusm esopotam icus) (Powell e ta l., 1997).
O nly very recent studies indicate that tw o or three form s o f G nR H are present in m am m als. T o date, HPLC-RIA or im m unocytochem ical data have been used to show that chicken G nR H -Il is expressed as a protein in the brains o f m onotrem es, m arsupials and placental m am m als including prim ates (D ellovade et al., 1993, King et al., 1989, 1990 and 1994, Lescheid et al., 1997). C hicken G nR H -II cD N A has been isolated and sequenced from m onkeys (Terasawa, personal com m unication) and hum ans (W hite et
al., 1998 and personal communication).
In all vertebrates the location o f m ultiple forms o f G nR H in individual brains shows a sim ilar pattern (Fig. 1-3). Most antisera cross-react with m ore than one form o f GnRH, but a few antisera have greater specificity for a single form o f GnRH. Antisera that are specific to cG nR H -Il have been used to show that G nR H -containing neurons in the m idbrain contain only cG nRH -lI in representatives o f five classes o f vertebrates (Conlon
all detectable form s o f G nRH have been sequenced are shown. Form s o f GnRH that have been detected using only RIA-H PLC methods are not shown, sb = sea bream GnRH; s = salmon G nR H ; c-II = chicken GnRH-II; cf = catfish GnRH. Num bers in each box represent teleost superorders or infradivisions: 1. O steoglossom orpha (bony
tongued fish), 2. Elopom orpha (eels), 3. C lupeom orpha (herring), 4.
Protacanthopterygii (salm onids), 5. Ostariophysi (includes C haraciform es), 6. Stenopterygii (e.g. lightfish and dragon fish), 7. S copelom orpha (e.g. lantern fish and greeneyes), 8. Paracanthopteiygii (e.g. cod and hake), 9. A therinom orpha (e.g. m edaka, m olly), 10. Percomorpha (O rder Perciform es). D iagram m odified from Powell e ta l., 1997.
SEA BREAM Sb,
s,
C-II 10. TILAPIA Sb,s.
C-II H. BURTONI Sb, S, C-II PACU Sb, S, C-II GOLDFISH S, C-II CATFISH Cf, C-II 1. 4. SALMON S, C-IIFig. 1-3. Diagram showing location o f different G nR H forms in brain sections. The m odified parasaggital brain sections illustrate the location o f GnRH in tunicate (M ackie,1995), lam prey (Nozaki e ta l., 1984), shark (Nozaki e ta l., 1984), sturgeon (L epretre e ta l., 1993), salm on (Am ano et al., 1991), cichlid (W hite, S.A. et a i , 1995), new t (M uske,1993), chicken (M illam et al., 1993) and m usk shrew (Rissm an et al.,
1995). The relative location o f cell bodies expressing distinct forms o f G nRH for six different vertebrate classes are shown. In species in which differential location has not been reported, the same symbol is used fo r both forms o f GnRH: tunicate GnRH-1 and -II ( ^ ), lamprey G nR H -I and -III ( □ ) and dogfish G nR H -I and cGnRH-11 ( ^ ) in the respective brains. In species in which differential location has been reported the sym bols are mamm alian G nR H ( H ), cG nRH -I ( ♦ ), salm on G nRH ( A ), sea bream G nR H ( ^ ), and chicken G nRH -II ( • ).
T u n i c a t e L a m p r e y O S h a r k S t u r g e o n S a l m o n C l c h l l d N e w t C h i c k e n M u s k S h r e w
T he neurites, however, extend to m any areas of the brain. The second form o f G nR H (e.g. m OnRH, gpGnRH, cG nRH -I, sGnRH o r cfGnRH) is in neurons o f the preoptic- septal region, basal forebrain and para-olfactory areas. This localization pattern fo r two form s o f GnRH within a single brain has been confirm ed with specific cDNA probes in catfish (Bogerd et al., 1993 and Zandbergen e ta l., 1995).
In two cichlid species {H. burtoni and tilipia) with three forms o f G nR H , specific cD N A probes have been used to distinguish the location o f the three forms o f GnRH: sG nR H is in the olfactory-forebrain areas; sbG nRH is in the preoptic region; and cG nR H -II is in the midbrain region (W hite, S .A . et al., 1995, Parhar, 1997) (Fig. 1-3).
M ultiple forms o f G nRH in a single species is an ancient pattern in evolution as show n by the two distinct form s o f GnRH in a protochordate (Fig. 1-1) (Powell et al., 1996). Im m unocytochemical studies o f G nR H in several tunicate species show that G nR H is present in the brain (neural ganglion) (B ollner e t al., 1997) and in a neural plexus outside of the ganglion (Georges and Dubois, 1980, Powell e ta l., 1996). T o date, the two types o f G nRH have not been distinguished in two different populations of cells in this tunicate.
III. G nRH cDNA transcripts.
Each GnRH m RNA encodes a preproprotein with four regions: 1)- a signal region, 2)- the G nRH decapeptide, 3)- a glycine-lysine-arginine processing site and 4)- a GnR H -associated peptide (GAP). The m RNAs encoding the distinct members o f the G nR H family have the sam e four regions, but are obviously different in the region that encodes the horm one and even more variable in the region encoding the G A P (Fig. 1-4). T he G A P moiety is thought to hold the preproprotein in the correct conform ation for cleavage o f the m ature hormone.
The cDNA encoding the mGnRH precursor has been characterized for human (Seeburg and Adelman, 1984 and Adelman e t al., 1986), rat (Adelman e ta l., 1986,
Fig. 1-4. A com parison o f known preproGnRHs at the am ino acid level. A line (I) indicates that the am ino acid below matches the am ino acid in hum an mGnRH. The single letter code is used to designate amino acids. The preproG nR H is shown for seven types o f G nR H : m am m alian GnRH (human, rat, m ouse, frog); guinea pig GnRH (guinea pig); chicken G nR H -I (chicken); seabream GnRH (gilthead seabream , cichlid {H.
burtoni) and striped bass); chicken G nRH -II (tree shrew to goldfish); catfish GnRH-I
mGnRH gpGnRH cGnRH-l sbGnRH HUMAN RAT MOOSE TREE SHREW FROG GUINEA P I G CHICKEN GB.SEABREAM C IC H L ID S T R IP E D BASS TREE SHREW CA T FISH GH.SEABREAM M— KPIQKLLAGLILLTW CVE6CSS I I I I I I I I I n i i i M— ET IPXLHAAW LLTVCLEGCS S I M l I I I I I I m i l M 1LKIM AGILLLTVCLEGCS S I I m m i l I I I I M --ELVPKFLAGLILLTLCVGGCYA I I I I I M -K AFPTFALLFLVLLP-SAHVSDA I I H i m I I I I I I I M— GLIPKLLAGLVLLTLCVENGSG I M i l l I M— EKSRKILVGVLLFTASAAICLA I I MAPQTSNLWILLLLVWMMMSQGCC I I I HAAKILALWLLLAGTVF PQGCC I I I MAPQTFALWLLLVGTLL GQGCC I I I I I
MAS SHLGFLLLLLLIMAAHP GP SEA
I I I I I I M-VSVCRLLLVAALLLCLQAQLSFS I I I I M— CVSRLVLLLGLLLCVGAQLSNG cGnRH-l! I I I I C IC H L ID M— CVSRLALLLGLLLCVGAQLSFA I I I I S T R IP E D BA SS M— CVSRLVLLF6LLLCVGAQLSNA G O L D F IS H -1 6 0 L D F I S B - I I M-VHICRLFW MGMLLCLSAQFASS I I I I I M -V BICR U TV H G K U 1FI.SV Q FA SS QHWSYGLRPG GKR m i i i i i i i I I I QHWSYGLRPG GKR i i m i i i i i I I I QHWSYGLRPG GKR l l l l l l l l l l I I I QHWSYGLRPG l l l l l l l l l l GKR l l l l l l l l l l QHWSYGLRPG GKR 1 n i l I I I QYWSYGVRPG I I I GKR l l l l l l l I I I I I QHWSYCLQPG GKR l l l l l l l I I i l l QHWSY6LSPG GKR l l l l l l l I I I I I QHWSY6LSPG GKR l l l l l l l I I I I I QHWSYGLSPG GKR n i l I I I I I QHWSHGWYPG GKR n i l 1 I I I I I QHWSHGWYPG GKR QHWSHGWYPG GKR n i l 1 I I I I I QHWSHGWYPG GKR n i l 1 11 I I I QHWSHGWYPG GKR m i l I I I I I QHWSHGWYPG GKR n i l 1 I I I I I QHWSHGWYPG GKR
GKR D A EM LIDSPQ EIVK EVG QLA ETQ BFECTXaQ PRSPLRDU ^GArESLIEEETGO--- K K I
GKR DTESLODM YHETPNEVAIFPELERLECSVPQSR— LNVLRGAIMMWLEGENR--- K K I
I I M i l l I I M i l I M M M M M M M I M M M M U M M l
NIEPLVDSFQEM AKEIDQLAEPQHTECTLBQPRSPLRDLKGAIESLM EEETGQ--- K K I
M M M i l l I I I I I I I M l M M I I I KAENLVESFQEIANEHESLGEGQKAECPGSYQBPRLSDLKETMASLIEGEARR--- K E I I I M i l l I I I I I I M M I M M M M M M M i l l I I NXEBLVDSFQEM GKEEDQHAEPQIIFECTVBHPRSPLRDIRGAIERLIEEEAGQ-I GKEEDQHAEPQIIFECTVBHPRSPLRDIRGAIERLIEEEAGQ-I M M M l I I I I M M I M M M M I I I M I M M I NIEBLVESFQEM GKEVDQHAEPQBFECTVBVIPRSPLROLRGAI£SLIEEEARQ-M GKEVDQHAEPQBFECTVBVIPRSPLROLRGAI£SLIEEEARQ-M GKEVDQHAEPQBFECTVBVIPRSPLROLRGAI£SLIEEEARQ-M GKEVDQHAEPQBFECTVBVIPRSPLROLRGAI£SLIEEEARQ-M GKEVDQHAEPQBFECTVBVIPRSPLROLRGAI£SLIEEEARQ-M I I I M M I M M M M M I M M M M M M M I
K A E N LID SFQ EIA K EA D Q IA EPQ B FEC TISQ PR SPLR A LK G A LESLIEEE IG Q
-I -I -I I I I I I I I I I I I M l I I I I - KKM I t -KKM M l - K K I M l I I I I I I I D U JS L SD T L G N IIE B FPB V D S PCSVIGCVEEPBVPRMYRMKGFIGSERDIGHRMYKK-I I I I I I I DLDNFSDTLGNMVEEFPRVEA PCSVFGCAEESPFAKMYRVKGLLASVAERKMDTGBSRNERFL I I I I I I I E ID G L SE T L G -IV G G FPB V E T PCRVtGCAVESPFPKIYRMKGDAVTDRENGPRTYKK-I I I ASNSPQDPQSALRPPAPSAAQTABSFRSAAIJVSPEDSVPHEGRTIAGHSLBRKQHLMRTLI.SAAGAPR. I I I I I I I I I E ID S Ï S S P E IS G E I K L - C E A G ---EC SY L R P L R T N IL K -S IL ID T L A R K F Q --- KRK I I I I I I I I E L D S F G T S E IS E E IK L -C E A G ---E C SY L TPOBRSVLR-NILLDALARELQ--- KRK I I I I I I I I E L D S F G T S E IS E E IK L -C E A G --- EC SY L--- R P Q R R S IL R -N IL ID A L A R E L O ---KRK I I I I I I I E L D S F G T S E IS E E IK L -C E A G --- EC SY L--- RPQ RR RV LR-N IILD A LA RELQ ---KRK I I I I I I
EID V Y D SSEV SG EIK L -C EA G --- KCSYL--- R P Q G R N IL K -T IL U D A IIR D S Q ---KRK
I I I I I I
CfGnRH
c
I I I I C A T F I S H - I I M— GIKRALW HM W CVW LQV-SA sGnRH mRNA2 sGnRH mRNA1 --- GH. SEABREAM 1 I I I 1 M— EASSRVTVQVLLLALW QVTLS 1 I I 1 1 M -“ E AS SRVT VQVLLLALWQVTLS I I I I 1 M - - E AGSRVIMQVLLLALW QVTLS I I I t 1M— RPYNV IW M W LLA LVLHAV LS RED SEABREAM C IC H L ID MIDSHIPMAN GO LD FISH M - -EGKGRVLVQLLMLACVLEVSLC SOCKEYE I I 1 M— DLSNRTW QVW LALVAQVTLS ATLANTIC M— DLSNRTW QVW LALVAQVTLS CHINOOK VRWVLALVAGVTLS RAINBOW TROUT 1 1 1 M --D LSN R TW Q V W LA LV A Q V TLS BROOK TROUT M— DLSNRTW QVW LALVAQVTLS --- BROWN TROUT M— DLSNRTW QVW LALVAQVTLS
--- SOCKEYE
I I I 1
M— D L SSK T W Q W H L A L IA Q V T PS RAINBOW TROUT M - -D L S SKTFVQW MLALIAQVTFS _____ MASU M— DLS SKT W Q W M LA LIA Q V TFS n i l I I I I I I I QHWSBGLNPG GKR l i n n I I I I I QHWSYGWLP6 GKR QHWSYGWIfG GKR l i n n I I I I I QHWSYGWLPG GKR n i i i i I I I I I QHWSYGWLPG GKR n n i i I I I I I QHWSYGWLPG GKR n u n I I i n QHWSYGWLPG GKR l i n n I I 1 1 1 QHWSYGWLPG GKR l i n n I I I I I QHWSYGWLPG GKR l i n n I I I I I QHWSYGWLPG GKR l i n n n 1 1 1 QHWSYGWLPG GKR i n i n n 111 QHWSYGWLPG GKR l i n n I I 111 QHWSYGWLPG GKR l i n n I I I I I QHWSYGWLPG GKR m i l l I I I I I QHWSYGWLPG GKR SVGELEATIRMMGTGGWSLPEEASAQTQERLMÏNV1KDDSSB---I I SVGELEATIRMMGTGGWSUDEAMAQIQERLBPYNIINDDSSB---I I I I SVGELEATIBMMGTGGWSLPEETSAQTQERURPYNIINDGGY---I I I I GKR SVGEVEATFRMMDSGDAVLSIPMDSPM— ERLSPIBIVSEVDAEGLPLKEQR--- FPNRRGRD— I I I I I AVMQESAEEIPRSSGYLCDYVAVSPGNKPFRLKDLI,TPVAG R E I E E -I I I
S VGELE AT IBM M GTGGW SLPEEAS AQTQERLRP YNVIKDDS S P
---I I I I I FDRKKRFPNK I I FDRKKRFPNK I I ■FDBKKRFPNN I I ■FNRKKRFFHE I i l I I SVGELEATIKM M DTGGW ALPEETSAHVSERLRPYDVIL---KKMMPHK I I I I I SVGELEATIKM M DTGGW ALPEETSAHVSERLRPYDVIL---KKHMPHK I I I I I SV 6ELEATINM M DTGGW ALPEETSAHVSERLRPYDVIL---KKWŒPHK I I I I I SVGELEATIKM M DTGGVW LPEETSAHVSERLRPYDVIL---KKHMPHK I I I I I
SVGELEATIKM M DTGGW A LPEETSAH VSERLRPÏD VIL---KKHMPHK
I 1 1 I I SVGELEATIKM M DTGGW ALPEETSAHFSERLRPYDVIL--- KKHMPHK I I SVGELEATIRMMDTGGVMALPEETDAHIPERLRPYDVNG--- C H IN K E L I I I I SVGELEATIWWDTGGVMALPEETGAHIPERLRPYDVMS--- KKRMPHK I I I I SVGELEATIPMMDTGGVMALPEETGAHIPERLRPYDVMS--- KKRMPHK On
A z z d c t a L , 1991 and O ikaw a er a/., 1990), m ouse (Mason ei ai., 1986), tree shrew (Tupaia glis belangeri) (Kasten e t al., 1996) and frog (Hayes et al., 1994) (Fig. 1-4). An in situ hybridization study using a probe to the m G nR H cDNA confirmed that the
mRNA in a mamm al (m ouse) is present in the preoptic area o f the brain (see Seeburg et
al., 1987).
T he cDNA encoding a novel form o f GnRH has been isolated from the hypothalam us o f guinea pig (Cavia cobaya) (Jim enez-Linan et al., 1997) and has not been reported in any other species to date (Fig. 1-4). T he G nRH decapeptide differs from its m am m alian counterparts by two am ino acid substitutions in positions 2 and 7. These two am ino acid substitutions are unique in com parison to all other known forms o f GnRH.
O nly one cDNA for cG nRH -I has been identified. The cG nRH-I cD N A was isolated from chicken brains (Dunn et al., 1993), but in situ hybridization has not been reported.
Seabream GnRH cD N A has been isolated from gilthead sea bream, cichlid and striped bass (G othilf e ta l., 1995, W hite, S.A. e ta l., 1995 and Chow e ta l., 1998). This fom i o f GnRH represents the third form o f G nR H in several species o f fish and hence the use o f a specific cDNA probe to localize the neurons that express sbGnRH was im portant. In two cichlids (H. burtoni and tilipia), the sbG nRH was expressed in neurons in the preoptic area, whereas sGnRH was in the rostral forebrain and cG nR H -II was in the m idbrain (W hite, S.A. e ta l., 1995 and Parhar, 1997) (Fig. 1-4).
cD N A s for cG nRH-II have been isolated from m onkey (Terasawa, personal
com m unication), tree shrew (Kasten e ta l., 1996), catfish (Clarias gariepinus) (B ogerd
et al., 1993), gilthead seabream (G othilf et al., 1996), cichlid (W hite et al., 1994) and
striped bass (Chow e ta l., 1998). An interesting aspect o f m olecular phylogeny is that goldfish are tetraploid and hence two cD NAs encoding the cG nRH-II precursor were isolated (Lin and Peter, 1997). Specific probes fo r cGnRH-II have been used to definitively identify this form o f G nR H in the midbrain of tree shrew (Kasten et a i .
1996), catfish (Zandbergen ct al., 1995), two cichlids (W hite, S.A. ct al., 1995 and Parhar, 1997) and the gilthead seabream (G othilf et a i , 1996).
T w o cD N A s encoding catfish GnRH (cfGnRH) were isolated from African catfish (C. gariepinus) (Bogerd et ai., 1993). The two cDNAs encode the identical horm one, but differ by four nucleotide substitutions that lead to the change o f two am ino acids in the G A P region. In situ hybridization has localized the cfGnRH form to the rostroventral forebrain and preoptic region (Zandbergen et ai., 1995).
T he cD N A encoding sGnRH mRNA2 has been sequenced for six salmonids
(K lungland e t al., 1992 a and b, Ashihara et al., 1995, C oe et al., 1995), three perciform s (a cichlid and both red and gilthead sea bream ) (Bond e t ai., 1991, O kuzaw a et al., 1995 and G othilf et al., 1996), one batrachoidiform (m idshipman) (G rober e ta l., 1995) and one cyprinid species (goldfish) (Lin and Peter, 1996) (Fig. 1-4). In situ hybridization studies done on the brains o f different salmonids detected cell bodies with sGnRH from the olfactory bulb to the preoptic area o f the brain (Suzuki et ai., 1992, Bailhache et ai., 1994).
D ue to the tetraploidization o f the salmonids, additional cD NAs that encode sGnRH m R N A l have been characterized for sockeye salmon (A shihara et ai., 1995), rainbow trout (C hapter 3) and masou salm on (Suzuki et ai., 1992). T he portion o f the m R N A l encoding the horm one has two nucleotide substitutions in com parison to the sGnRH m R N A 2 transcripts, but because these changes occur in the wobble position o f the codons for am ino acids 2 and 6, the sGnRH decapeptides translated from these two different sG nRH -encoding m R N A s are the same. Differences am ong the signal and G A P regions of the sGnRH m R N A l-encoded products put them in their own separate class from their sGnRH m RNA2 counterparts (Fig. 1-4).
To trace the evolution o f G nR H in detail requires more inform ation than provided by the ten am ino acid GnRH peptide. The precursor m olecule o f each peptide as deduced from the cD N A s is useful in preparing an evolutionary map. V eiy little conservation o f
the G A P region o f the precursors is observed (Fig. 1-4), but this region m ay be crucial for determ ining the diversification o f the GnRH family in finer detail. F or example, in the mGnRH, gpG nR H and cG nR H -I precursors, the G A P am ino acids are closer in identity than between mGnRH and any other G nRH precursor (Fig. 1-4), confirm ing the idea that cG nRH -I was derived from mGnRH relatively recently. In the salm onid and human GAPs, how ever, only 4-9% identity is found, but the secondary structural level has som e conservation. A central cysteine can be aligned in each G A P for mGnRH, gpGnRH, cG nRH -I, cG nR H -II (except tree shrew) and sbG nRH , but not fo r sGnRH (Fig. 1-4). Also, cG nR H -II has a second cysteine that can be aligned with cfGnRH's cysteine. Although speculative, I suggest cfG nRH may be derived from a duplication of the cG nRH -II gene. Am ino acid identity is not helpful here as cfG nR H is 2 amino acids different from either m G nRH o r cG nRH-II. Furtherm ore, a leucine in m GnRH (position 5 o f G A P) is repeated in cG nRH -I, sbGnRH (except cichlid) and sGnRH (except
goldfish), but not in cG nRH -II o r cfGnRH. The sum o f evidence supports the idea that m GnRH was ancestral to salmon GnRH, seabream GnRH and chicken G nR H -I. The accuracy o f determ ining the lineage o f G nRH forms will increase as more cD N A s are sequenced and the G A P regions can be com pared for shorter evolutionary distances.
IV. G nRH genes in vertebrates.
Five different genes that encode G nRH in vertebrates have been identified to date; each o f these genes encodes only one form o f GnRH: m G nR H , sbGnRH, cGnRH-1, cG nR H -Il or sGnRH (Fig. 1-5). T he gene from which m G nR H is transcribed has been identified in human (Adelman et al., 1986, H ayflick et at., 1989 and Radovick et a l,
1990), rat (Adelm an e ta l., 1986, Bond et al., 1989 and Kepa e ta l., 1992) and m ouse (M ason et a i , 1986), whereas the cGnRH-1 gene was identified in chicken (Dunn et al..
Fig. 1-5. Com parison o f the G nRH genes fo r five distinct forms o f GnRH: m am m alian G nRH (m GnRH), chicken GnRH-I (cGnRH-I), seabream GnRH (sbGnRH), salm on G nR H (sG nR H ) and chicken GnRH-II (cGnRH-II). The sGnRH gene has been isolated and sequenced from an Atlantic salmon {Salmo salar) an d from a Pacific salmon
(O ncorhynchus nerka, sockeye salmon). T he cG nRH-I and sbGnRH gene has been
isolated only from chicken and striped bass, respectively, but the m G nR H gene has been characterized from m ouse, rat and human. T he cG nR H -II gene has been isolated in both striped bass and human. Exons I, II, III and IV are labeled. Introns are shown as lines. T he coding for the signal peptide is stippled, the horm one is black, the processing site and G nRH -associated peptide (GAP) are w hite, and the 5'- and 3'-untranslated regions are shown by diagonal lines.
I m G nRH RAT/MOUSE CHICKEN cGnRH-1 S b G n R H STRIPED BASS PACIFIC SALMON sG nR H ATLANTIC SALMON III IV
---
7^
m
m
/ / / / / / / / A
m
1
/ / / / / / / / A
u y / / / / / / / w -HUMAN cG nRH -II STRIPED BASS1993) and the salmon G nR H gene from two species o f salmon: an Atlantic salmon
(Salmo salar) (Klungland et at., 1992b) and a Pacific salm on (O ncorhynchus nerka)
(Coe e ta l., 1995). The genes encoding sbGnRH and cGnRH-11 were recently
characterized in striped bass (Chow et al., 1998). The cG nR H -ll-encoding gene has been isolated in monkey and human, but the sequences have not yet been published
(Terasawa, personal com m unication and W hite et al., 1998). The architecture o f the genes encoding the five different GnRH forms is the same with a com m on organization o f four exons separated by three introns (Fig. 1-5). In each gene, exon 1 encodes the 5'- untranslated region (5'UTR); exon 2 has the signal peptide, G nRH, a proteolytic
cleavage site and the am ino term inus o f a GnR H -associated peptide (GA P); exon 3 has the central GAP moiety; and exon 4 has the carboxy term inus o f G A P and the 3'- untranslated region (3'UTR).
Several interesting patterns are apparent in the regulatory 5'-flanking regions o f GnRH. The 5'-regulatory regions of the genes encoding mGnRH (R adovick et al., 1990, Kepa et al., 1992) and cGnRH-1 (Dunn et al., 1993) differ from the fish genes encoding sGnRH (Klungland et al., 1993, Coe et al., 1995) because the form er contain obvious regions o f alternating purines (adenine or guanine) and pyrim idines (thym ine or cytosine) o r blocks o f polydeoxyadenosine residues. M any o f the A T-rich regions in each gene may serve as targets for mem bers o f the Pit-l/O ct-l/U nc-1 (PO U ) fam ily o f transcription factors (Bendall e ta l., 1993) for m odulation o f G nR H expression.
Interestingly, one footprint in the proxim al prom oter region contains an A T-rich elem ent that has been implicated as a target for protein kinase C (PK C )-m ediated repressor activity resulting in dow nregulation o f the rat G nRH prom oter (Eraly and M ellon, 1995). Furtherm ore, a distal enhancer region within the rat GnRH prom oter also contains AT- rich elem ents that exert stim ulatory (Clark and Mellon, 1995) and inhibitory (Belsham et
Although much is known about the presence and distribution o f multiple form s o f G nR H within the brain, relatively little is known about their regulation. Two o r m ore G nR H genes within a single species indicate that each gene probably serves a distinct task. T he high degree o f sequence variability and lack o f obvious regions o f sequence conservation in the prom oters o f G nR H -encoding genes between species points to differences in the regulation o f each gene. A com parison o f the prom oters from known G nR H -expressing genes and the factors that control their transcription will be discussed in an upcom ing chapter o f this thesis.
V. GnRH function.
A num ber o f studies using im m unocytochem ical m ethods and characterization o f am ino acids and cD N A s all docum ent the existence o f m ultiple form s o f G nR H within the brain. However, very little is known about the distinct functions for individual GnRHs. Location o f each peptide m ay give a clue about function. For exam ple, close proxim ity o f G nR H -containing neurons to the pituitary implies a prim ary role as a gonadotropin releaser; detection in the forebrain, near olfactory nerve and/or retinal projections, implies a role in integrating olfactory and visual cues involved in reproductive behaviour; discovery in the m idbrain o f GnRH neurons with axons that term inate within the brain o r spinal cord im plies neurom odulatory roles possibly m odifying m otor-associated reproductive behaviours.
The difference in distribution o f the vertebrate GnRHs is attributable to at least two distinct em bryonic origins: one in the olfactory placode, which gives rise to the term inal nerve-septo-preoptic system, and one in the m idbrain, which gives rise to the posterior cG nR H -II system (Muske, 1993). The em bryonic origin of a third form o f G nR H in the brain, at least for sbGnRH, is also the olfactory placode. The distribution gives a clue as to the functions o f the forms o f GnRH.
A. R elease o f g o n a d o tro p in . One clear function o f G nR H is to activate the synthesis and release o f gonadotropins from the pituitary. Neurons that contain im m unoreactive m G nR H , cG nRH-I, sGnRH, cfGnRH o r sbG nRH have been localized in the basal forebrain-septo-preoptic system and their axons shown to term inate on the hypothalam o-hypophysial portal vessels in tetrapods or in the pituitary in teleost fish. This location suggests that these GnRHs are predom inantly involved in reproductive control as gonadotropin-releasing factors. Teleost bony fish offer a unique insight into which form o f G nRH is the gonadotropin releaser because the neurites containing GnRH end directly in the pituitary. The num ber o f forms o f G nR H found in term inals in the pituitary, how ever, varies with the species. In fish with three forms o f G nR H , the pituitary o f the cichlid contained only sbGnRH (Powell et al., 1995), but the pituitary of the sea bream fish (Powell et at., 1994) and tilapia (Parhar, 1997) contained both sbG nRH and cG nR H -II. In fish with two forms o f G nR H , the pattern also varies: goldfish (Kim e t al., 1995 and Yu et al., 1988) have two form s o f G nR H (sGnRH and cG nR H -II) in the pituitary; eels and catfish have predom inantly one form o f GnRH (m GnRH o r cfG nR H , respectively) in the pituitary, but very small am ounts o f cG nRH -lI also have been detected (M ontero e ta l., 1995 and Zandbergen e ta l., 1995); salmon have only sG nR H (A m ano et al., 1991). Therefore, m ore than one form o f G nR H m ay act on the pituitary. Physiologically, all three forms o f GnRH, if given exogenously, release gonadotropin in sea bream (Zohar et al., 1995) and both form s release gonadotropin o r induce ovulation in the goldfish (Habibi e ta l., 1992).
The (H yp9)m G nRH form, which is found in mam m als and am phibians, binds to m G nR H -like pituitary receptors to stimulate the release o f LH and FSH in vitro and in
vivo in a sim ilar dose-dependent but less potent m anner than m GnRH (Gautron et
a /.,1992). In m am m als and amphibians, post-translational enzym atic m odification of existing G nR H m olecules, rather than gene duplication and subsequent em ergence o f
novel GnRH forms, m ay be an alternate strategy to achieve m olecular heterogeneity and diversity o f function.
B. R elease o f g ro w th h o rm o n e a n d p ro lac tin . GnRH in vertebrates releases gonadotropins from the pituitary, but in some species o f fish, G nR H also releases growth hormone (M archant e ta l., 1989 and M elam ed e ta l., 1995) and prolactin from the pituitary (W eber et a i , 1997). T he three native forms o f G nRH in tilapia brain stimulate the in vitro release o f prolactin from the pituitary with the follow ing order o f potency: cG nRH -II>sG nRH >sbG nR H . This effect was shown to be a direct effect on the prolactin cells in the pituitary. It is not surprising that G nRH affects both growth horm one and prolactin as these two hormones are m em bers o f the same protein family and are thought to have resulted from an earlier gene duplication. C om parable doses o f GnRH were effective in stimulating the release o f gonadotropin, grow th horm one and prolactin from pituitary fragments o f fish (W eber et al., 1997). In addition, the order o f potency for the three types o f GnRH for release o f gonadotropin in sea bream (Zohar et
al., 1995) was the same as for release o f prolactin in tilapia (W eber et al., 1997).
C . P u ta tiv e effects on b eh a v io u r. The localization o f cG nR H -II in m idbrain neurons in all vertebrates including prim ates (Lescheid et al., 1997) indicates this GnRH form may serve a neurom odulatory role in reproduction and behaviour. Axons
containing cGnRH-II extend to many regions of the brain (Lepretre e ta l., 1993 and Muske, 1993) and are known to descend in the spinal cord (M illar and K riebel, 1986, W right and Demski, 1991). Indirect evidence shows that cG nR H -II is the form that can elicit the late, slow excitatory-postsynaptic potential in frog sym pathetic ganglia (Jones,
1987). In addition, the application o f GnRH to the m idbrain resulted in changes in reproductive behaviour (Pfaff, 1973). The early expression o f cG nRH -II in fish
em bryos (W hite, R.B. e ta l., 1995) and fetal m onkeys (Quanbeck e ta l., 1997) suggests that additional roles are important.
D. Local action in re p ro d u c tiv e o rg a n s. G nR H is im portant in the developm ent
and m aturation o f reproductive tissues o f vertebrates. In m am m als G nR H has been utilized to serve new roles as dem onstrated by its postulated involvem ent in the
developm ent o f the placenta (Kelly e ta l., 1991) and in em bryo im plantation and growth (Li et al., 1993). A comm on them e exists for GnRH transcripts expressed in human placenta, mammary gland, ovary and testis whereby the 5'-untranslated region (UTR) is longer than in brain transcripts (Dong et al., 1993). T he elongated 5'-UTRs can result from use o f a different upstream start site com pared to brain transcription (Dong e ta l.,
1993) o r to the retention o f intron 1 in the m RNA (Seeburg and A delm an, 1984, Radovick e ta l., 1990, D ong e ta l., 1993). The long 5'-UTRs in placental, breast and gonadal tissue com pared with brain indicates differential regulatory m echanism s. To date, only prelim inary data exists on G nR H expressed in the gonads o f low er
vertebrates, but GnRH-like activity and/or binding sites are present in the ovary o f goldfish, carp and African catfish (see Habibi et al., 1994). sGnRH m RN A has also been detected in m idshipm an ovary and testis by Northern analysis (G rober et al., 1995) and both sGnRH and cG nRH -II cDNAs have been isolated from the goldfish ovary (Lin and Peter, 1996). Finally, the vestiges of the early role o f G nRH in gonadal developm ent are still seen in the tunicate where gonadogenesis fails to occur in the absence o f the nerve plexus containing G nRH (see Irons, 1986).
E. F u n ctio n s in a n c e stra l an im als. In a protochordate, the injection o f tunicate(t) G nR H -I or tGnRH-II m onom ers into the visceral blood sinus o f adult tunicates results in a m easurable increase in estradiol content o f the gonads within 6 hours (C raig et al.,
1997). This observation as well as im m unocytochem ical studies show ing num erous ir- G nR H neurons surrounding adult and juvenile gonads and gonoducts, suggest tOnRH
peptides are released into the bloodstream to act directly on the gonads. Current evidence suggests that it is unlikely tunicates contain an organ hom ologous to the pituitary. Therefore, their neurohormones m ay act directly on the gonads or other tissues. T he observation that tunicate G nRH s affect the estradiol content o f the gonad provides evidence for the early existence o f G nR H receptors.
The structure o f a molluscan GnRH-like substance has not been reported, but an extract from ganglia of H. trivolvis released gonadotropins from dispersed goldfish pituitary cells (Goldberg et at., 1993). T his suggests that the G nR H fam ily m ay be w idely distributed.
VI. Purpose o f thesis.
Pacific salm on are an important biological resource. T hey provide food for over 22. species o f m am m als and birds and act as a huge nitrogen and carbon sink for coastal w atersheds. T heir dramatic life cycle and incom pletely understood m igratory habits provide a m ystique that still permeates the culture o f coastal aboriginal peoples.
I chose both sockeye salmon and rainbow trout to study because o f their differences in habitat and reproductive patterns. Sockeye salm on spend the bulk o f their adult lives in the ocean and then return to spawn in natal freshw ater stream s (anadrom ous) and die (sem elparous). In contrast, rainbow trout spend their entire lives in freshw ater lakes and stream s and can spawn more than once (iteroparous). Som e rainbow trout spawn in spring, w hereas others spawn in the fall. A tlantic salm on, which are anadrom ous like the sockeye salm on, can spawn, return to sea and spawn again, som etim es as m any as four tim es in a lifetime.
Reproduction is dependent on external cues such as water tem perature, light and the behaviour o f conspecifics. Age, fat content and the physiological m ilieu also play an im portant role in the readiness o f a fish to m ate and reproduce. M uch, if not all, o f this
synthesis and secretion o f GnRH, which is central to reproduction. This thesis examines the control and location o f expression o f GnRH in these two contrasting fish species to im prove our understanding o f the mechanisms involved that integrate the events governing reproductive processes.
The m ain purpose o f this thesis is to study the regulation and expression o f GnRH in the brain and reproductive tissues o f sockeye salm on and rainbow trout. The first objective was to isolate and characterize the gene and cD N A that encodes sGnRH in a salm onid. I isolated the first gene encoding sGnRH from the tetraploid sockeye salmon. This particular gene is referred to as the sGnRH gene2. A t the time this was achieved only the gene encoding m G nRH and cG nRH -I had been established and characterization o f the gene encoding sG nR H m RNA2 in Atlantic salm on had ju st been com pleted. Characterization o f the sG nR H gene provides the basis to determ ine which regions of this particular gene are conserved after the duplication o f the entire genom e as occurred in salm on approxim ately 27 m illion years ago.
For m y second objective the start site used to generate the sGnRH mRNA2 transcripts in the brain was determ ined unequivocally by prim er extension analysis. Also a Southern analysis was done on sockeye salm on genom ic DNA using a portion of the sGnRH gene2 as a probe to provide evidence that other genes encoding GnRH were also present in the salm onid genome.
My third objective was to sequence the 5'-flanking region o f the gene encoding sGnRH m R N A l from rainbow trout so I could com pare the regulatory regions for GnRH from rainbow trout, Pacific salmon (sockeye salm on) and Atlantic salmon. For example, a large segm ent o f 1152 nucleotides in the prom oter region o f the sockeye salmon sGnRH gene2 is m issing in comparison to the A tlantic salmon gene.
Furtherm ore, all three genes share strong conservation in the m ost proxim al 200 bps of their regulatory 5'-flanking DNA. However, except for 100 bps in the upstream 5'- flanking region, the sG nR H genel bears no resem blance to either o f the tw o genes that
encode sGnRH m RNA2. This indicates that all the sG nR H -encoding genes
characterized to date are governed by different upstream regulatory factors or that the response elem ents are organized in a different pattern so that the sG nR H genes share some com mon control elem ents.
Because G nR H has been shown to be present in m am m alian placenta, ovary and testis, salm onid retina and in amphibian sym pathetic ganglia, my fourth objective was to examine various salm onid tissues at different life stages to determ ine the expression pattern o f GnRH in tissues outside of the salm onid brain. My early w ork provided evidence that G nR H was indeed expressed as mRNA in the gonadal tissue o f 1.5-year- old sockeye salm on. This work was follow ed by the m onthly exam ination o f ovarian and testicular tissue from both sockeye salmon and rainbow trout d uring the first two years o f their lives.
These tissue expression studies revealed that the regulation o f G nR H in the brain and gonad o f salmonids differs. I found that sGnRH m RNA2 transcripts w ere generated for much shorter durations in the ovaries and testes o f sockeye salm on than in rainbow trout. For each of the tw o years that the reproductive tissues w ere extensively examined, sockeye salmon expressed G nRH mRNA for no more than one m onth (October) o f the first year and not at all in the second year, whereas rainbow trout expressed G nRH from M ay through O ctober o f the first year and in Decem ber o f the second year. Fish that mature precociously in year 2 in both species express GnRH in M ay and June (sockeye salmon) or from June to O ctober (rainbow trout), suggesting that reproductive maturity is coordinated with an increase in G nRH expression in the gonads.
My fifth research objective was to study the regulation of expression o f the sGnRH mRNA2 transcripts in the salm onid ovaries and testes. I found that the transcripts were generated from an alternative upstream prom oter that differed from that used in the brain. I determ ined from prim er extension analysis that an additional 322 nucleotides of 5'-untranslated region is included in the gonadal transcript, but not the brain transcript.
Also, intron I may be retained and when coupled to the alternative upstream start site results in sGnRH cDNA2s in the gonads with much longer 5'-U TR s than their brain counterparts.
A study o f the regulation and expression o f sGnRH m R N A 2 in both sockeye salmon and rainbow trout em bryonic tissue was m y sixth objective. Tissues collected every other day for rainbow trout 1 to 42 days after fertilization (D A F) and sockeye salmon 1 to 68 D A F were examined by reverse transcriptase-polym erase chain reactions. sGnRH m R N A 2 was expressed in rainbow trout and sockeye salm on tissue beginning from 10 and 3 0 D A F, respectively. U se o f the upstream alternative prom oter for each species was tem poral, but was used consistently ju st prior to and follow ing hatching. Detection o f G nR H expression was not evident in gonadal tissue follow ing this developm ental period until four m onths later and only in rainbow trout.
As described above, transcripts encoding mGnRH, sG nRH and cG nR H -II have been isolated and characterized from the ovaries o f various species. H ow ever, to date, translation o f these m essages into protein has not been show n unequivocally. My seventh objective was to determ ine if G nR H expressed in the salm onid ovary and testis was translated into protein. I exam ined rainbow trout ovaries and testes by HPLC and com petitive RIA. RIA o f H PLC fractions eluted from extracts o f im m ature or
precocious (jill o r jack) 17-month-old ovaries and testes show ed weak cG nRH -II im m unoreactivity for only the jill ovaries. Unexpectedly, RIA detected strong
im m unoreactivity for four different forms o f GnRH in both the ovaries and testes o f 18- to 20-m onth-old rainbow trout that m atured precociously. H PLC elution and cross reactivity with specific antiserum indicated the presence o f m G nR H , cG nR H -II, sGnRH and an unknown GnRH in these tissues. Subsequent studies on norm al developing tissues gathered from 21-m onth-old fish did not detect G nR H im m unoreactivity. This indicates that detectable GnRH peptide is present only for a short period during the m id sum m er in precociously m ature rainbow trout.
