Detection of low plasma estradiol concentrations in nesting green turtles (chelonia mydas) by hplc/msms
Detection of Low Plasma EstradiolConcentrations in Nesting GreenTurtles (Chelonia mydas)by HPLC/Ms-MsI.Y. MAHMOUD1, A.Y. ALKINDI2, T. KHAN3, AND S.N. AL-BAHRY1Ã1Department of Biology, College of Science, Sultan Qaboos University, Al-Khod, Muscat,
2University of Nizwa, Nizwa, Sultanate of Oman3Central Analytical and Applied Research Facilities, College of Science, Sultan Qaboos University,
In previous studies on nesting green turtles under natural conditions from different geographical
regions, 17-b-estradiol (E2) was either undetectable or detected at very low levels. RIA and otherrelated techniques were not sensitive enough to measure low E2 values in the green turtles. In thisstudy, a sensitive method was used in detecting low hormone concentrations: high performanceliquid chromatography with tandem quadruple mass spectrometry (HPLC-MS/MS). Using thistechnique, estradiol for the first time was detected in nesting green turtles during the peak season(June–October) at Ras Al-Hadd Reserve, Oman. The E2 values recorded from this study were thehighest ever recorded from nesting green turtles in any geographical region, but the levels did notvary significantly throughout different phases of nesting. The presence of E2 during nestingpresumably plays a role in the physiology and behavior of this species. Ras Al-Hadd hosts one ofthe largest nesting populations of green turtles in the world, and an understanding of their nestingpatterns may be of value in conservation and management programs for this endangered species.
J. Exp. Zool. 315:170–174, 2011. & 2011 Wiley-Liss, Inc.
How to cite this article: Mahmoud IY, AlKindi AY, Khan T, Al-Bahry SN. 2011. Detection of lowplasma estradiol concentrations in nesting green turtles (Chelonia mydas) by HPLC/Ms-Ms.
Sea turtles in general have a multiannual reproductive cycle.
Our previous attempts to measure E2 levels in the green turtle
Following an internesting period of approximately 2–5 years,
population at Ras Al-Hadd during nesting using the standard
complete follicular growth and maturation occur. During the
Radioimmunoassay (RIA) technique were unsuccessful (Al-Habsi
nesting period, females usually lay from two to six clutches at
et al., 2006). We concluded that E2 is probably expressed in green
turtles during nesting but at such low levels that it cannot be
The chronological sequence of the ovarian cycle in sea turtles
commences with vitellogenesis in which 17-b-Estradiol (E2)
In this study, therefore, we have used a more sensitive method
levels rise significantly. This is followed by migration to the
to detect this hormone using high performance liquid chromato-
nesting grounds. Mating usually takes place near the nesting
graphy coupled with tandem quadrupole mass spectrometer
beaches and ovulation occurs just before nesting (Wibbels et al.,
’90; Miller, ’97; Owens, ’97).
Vitellogenesis in green turtles (Limpus, ’95) and in logger-
heads (Wibbels et al., ’90; Miller, ’97) occurs a few months before
ÃCorrespondence to: S.N. Al-Bahry, Department of Biology, College of
Science, Sultan Qaboos University, Al-Khod, Muscat, Sultanate of Oman.
main hormone triggering follicular growth. Ho (’87) reported that
Received 2 July 2010; Revised 3 December 2010; Accepted 7 December
E2 stimulates vitellogenesis in reptiles, and Owens (’76) observed
that E2 injections stimulate vitellogenesis and oviductal growth
Published online 11 January 2011 in Wiley Online Library (wileyonline
PLASMA ESTRADIOL IN NESTING GREEN TURTLES
We have applied this method to analyze plasma E2 levels in
green turtles at Ras Al-Hadd Reserve during nesting phases. WithHPLC-MS/MS, E2 can be detected and measured in the plasma ofnesting green turtles, and this technical capability may be ofimportance for better understanding their nesting patterns.
However, more research using this technology will be neededto evaluate E2 levels throughout the reproductive cycle.
Study SitesRas Al-Hadd Reserve is located at the eastern tip of Oman, at the
Figure 1. A calibration curve for determination of estradiol levels.
junction of the Gulf of Oman and the Arabian Sea at 591480E andbetween 221320N and 221140N. The Reserve consists of approxi-mately 20 beaches with lengths ranging from 50 m to 5.3 km.
E2 was separated using a High Performance Liquid Chroma-
tography system (Agilent 1100, Palo Alto, CA) with an XterraC18, 2.1 Â 50 mm, 3.5 mm column (Waters Corp., MA). Acetoni-
trile/water (70/30, v/v) was used as mobile phase (Sigma Aldrich,
The samples are categorized as follows: (1) Emergence, appear-
ance from sea; (2) Excavation, digging of the body and nest
Validation of the extraction procedure and HPLC-MS/MS
chambers; (3) Successful, completion of all phases including
method was achieved by comparing extraction recoveries. The
emergence, excavation, oviposition, and burying the nest; and (4)
peak area in the chromatogram of a 10 pg/mL standard E
Unsuccessful, return to the sea after excavation but without
solution in water/acetonitrile was compared with a standard, at
the same concentration, spiked turtle serum that had been carried
Ten milliliter of blood was taken from the cervical sinus from
through the protein precipitation extraction procedure. A blank
each nesting turtle within 5 min of capture to minimize stress
plasma sample was run and subtracted from the spiked sample.
(Owens and Ruiz, ’80). All samples were collected during the peak
The levels of estradiol in the blank were below the limit of
nesting season (June–October in 2005 and 2006). Fifty turtles
quantification for the method and this was used to back-calculate
were sampled including 9 during emergence, 11 excavations, 13
the concentration in the calibration curve standards and
successful, and 17 unsuccessful. Only one sample was taken from
subsequent samples. The efficiency of recovery was 98%. The
each turtle. The tag turtle number was recorded to avoid taking
back-calculated concentration from the protein precipitation
additional blood samples from the same turtle.
extraction method was used to evaluate the steroid concentrationin actual turtle serum samples.
A calibration curve was constructed in acetonitrile/water, first
For protein precipitation, a 500 mL aliquot of acetonitrile was
for validation only and then for the actual samples. It was
added to 500 mL of plasma (1:1 ratio by volume). The sample was
constructed using plasma that was spiked and extracted in the
vortex mixed for 5 min and then centrifuged at 3,000 rpm for
same way as the sample (Fig. 1). Back-calculated concentrations
10 min in an Eppendorf (Hamburg, Germany) microcentrifuge.
were obtained using a calibration curve over the concentration
The precipitated plasma protein which sedimented to the bottom
range 0.1–6,000 pg/mL with a correlation coefficient of 40.99.
of the microcentrifuge tube was discarded and the top layer
This gives an accurate indication of the actual steroid concentra-
containing the extract was analyzed by injecting a sample of
A 1 ng/mL standard solution of E2 (Sigma Aldrich, UK) was used
The Kruskal–Wallis test, a nonparametric test, was used to test the
A tandem quadrupole mass spectrometer (Quattro Ultima Pt,
equality of the population medians of the estradiol concentra-
Waters Corp., MA) was used in this analytical procedure. The cone
tions in the four nesting phases, whereas a one-way ANOVA was
voltage was set to 50 V and the collision energy was set to 18 eV.
used to test the equality of the population means for the four
The resolution settings were tuned to 0.7 Da at half height with ion
energies set to 1.0 V. The precursor ion for estradiol was set to271.2 m/z with the fragmented product ion set at 155.1 m/z.
Consequently, a multiple reaction monitoring experiment was run
with Argon collision gas at 2.8 Â 10À3 m Bar pressure. The estradiol
Results of E2 determinations for the four nesting phases, using the
was run in negative ion electrospray which generates [M-H]À.
HPLC-MS/MS method, are shown in Figure 2 and Table 1. The
mean concentration change of E2 (pg/mL) in nesting turtles was
In this study, E2 levels were the highest ever recorded in
highest in successful turtles, followed by emergence, unsuccess-
nesting green turtles in any geographic region. Moreover, E2
ful, and excavation (Table 1). However, the nonparametric
levels remained comparatively low throughout nesting with no
Kruskal–Wallis test shows that, even though the median
significant differences in values detected between the nesting
concentration is low in the excavation phase compared with
phases. It is unclear why E2 remains low and stable during
the other three phases, this difference is not statistically
nesting phases, but it may be that low E2 levels are sufficient to
significant (P 5 0.102). This result is confirmed by the use of a
maintain physiological and behavioral functions during nesting
one-way ANOVA test of equality of the population means
(P 5 0.128). Thus, our results indicate that there is no significant
Al-Habsi et al. (2006) investigated the plasma E2 levels in
change in the concentration of E2 throughout the nesting phases.
green turtles during different phases of nesting under natural
The means, medians, and ranges (95% confidence intervals) of E2
conditions at Ras Al-Hadd Oman. Using the RIA technique, their
concentrations (pg/mL) for the four nesting phases are summar-
results revealed that E2 levels were undetectable in all phases of
Similarly, Licht et al. (’79) and Lance et al. (’79) reported from
Florida that E2 levels during nesting phases in captive wild stockgreen turtles were mostly below the detectable levels of RIA
1072.0 pg/mL). On the other hand, the above two
Our results show that E2 concentrations can be reliably measured
investigators reported higher values in some of the captive
using the HPLC-MS/MS method, even when they are too low to
nesting turtles (approximately 50 pg/mL).
be detected by RIA or other comparable methods. Owing to its
Low or undetectable E2 levels during the nesting period are
extra sensitivity and enhanced selectivity, HPLC-MS/MS is a
common in sea turtles. Rostal et al. (’98) reported that in the
precise tool for detecting very low hormone levels during nesting
nesting Kemp’s Ridley sea turtles (Lipodochelys kempi) E2 levels
remained basal throughout the nesting season without anysignificant variation. Licht et al. (’82) reported that in Olive Ridleyturtles (Lipodochelys olivacea) E2 was quite low throughout thenesting season. Lance et al. (’79) suggested that duringinternesting before each ovulated clutch in the green turtles, E2is essential even at low levels for the completion of follicularmaturation.
In this study of wild green turtles from Ras Al-Hadd, the
values obtained were significantly higher than those observed incaptive turtles. One explanation for this discrepancy is thatcaptive turtles may not represent the true natural hormonevalues. Alternatively, the discrepancy may be owing to the use ofthe RIA technique vs. the HPLC-MS/MS method used here.
Despite the large number of mature follicles produced by
green turtles, E2 levels remain very low compared with otherchelonians, including other sea turtles. During vitellogenesis, E2
Figure 2. The 95% confidence intervals for the means of estradiol
concentrations rise significantly in sea turtles but decrease to
concentrations by nesting phases. Square crosses ( ) represent the
minimum levels for the rest of the reproductive cycle, including
means and the circled crosses ( ) represent the medians.
Table 1. Summary statistics of estradiol concentrations during nesting phases.
PLASMA ESTRADIOL IN NESTING GREEN TURTLES
Wibbels et al. (’90) also reported that E2 levels in the
then remains at a minimal value for the rest of the reproductive
loggerhead (Carreta carreta) rose significantly during the early
season of nesting, but declined drastically during the final phase
In summary, there is no significant difference among the four
of nesting. This indicates that the ovary may still be active during
phases, although the excavation phase showed appreciably lower
the early nesting period, undergoing vitellogenic activity
levels compared with other phases. Even though E2 was at a low
between successive clutches (Wibbels et al., ’90). However, in
level throughout nesting, it could be complemented by other
the green turtles at Ras Al-Hadd, the E2 levels remained
hormones, and this could be enough for nesting exercises to be
unchanged throughout the nesting season, which may indicate
carried out. The results of this study may be of value to the
that E2 levels are sufficient to carry on vitellogenesis between
overall understanding of the physiological and behavioral events
successive clutches. Wibbels et al. (’90) reported that during
of the nesting turtles at Ras Al-Hadd.
periods of vitellogenesis the E2 concentrations in the loggerheadrose significantly and remained high for approximately 4 weeks.
This suggests that E2 is the primary stimulus for vitellogenesis.
Just before migration, E2 levels decrease to a low level, whereas
the testosterone (T) concentration increases significantly. Thus, T
Al-Habsi AA, AlKindi AY, Mahmoud IY, Owens DW, Khan T, Al-Abri A.
may trigger migration (Owens, ’97). Moreover, in the loggerhead,
2006. Plasma hormone levels in the green turtle, Chelonia mydas,
E2 values were recorded during periods of vitellogenesis in nine
during peak period of nesting at Ras Al-Hadd-Oman. J Endocrinol
reproductively active females (tagged with a sonic transmitter),
which were bled repeatedly (five were bled before migration)
AlKindi AYA, Mahmoud IY, Al-Gheilani HM, Bakheit CS, Al-Habsi AA,
Al Kiyumi A. 2003. Comparative study of the nesting behaviour of
Guillette et al. (’91) reported that E2 levels during nesting in
the green turtle Chelonia mydas, during high- and low-density
the loggerhead ranged from 170 to 516 pg/mL, which is
nesting periods at Ras Al-Hadd Reserve, Oman. Chelonian Conserv
significantly higher than that of the green turtles. The different
values between the loggerhead and the green turtles may be
Guillette Jr LJ, Bojorndal KA, Bolten AB, Gross TS, Palmer BD,
Worthington BE, Matter JN. 1991. Plasma estradiol-17b, proges-
A possible explanation for low E2 levels reported here is
terone, prostaglandin F and prostaglandin E2 concentrations during
related to hormone-receptor affinity, as suggested by Porthe-
natural oviposition in the loggerhead turtle (Carretta caretta). Gen
Nibelle and Lahlou (’78), because different species have different
hormone-receptor affinities. High affinity receptors may require
Ho S. 1987. Endocrinology of vitellogenesis. In: Norris DO, Jones RE,
lower hormone concentrations for activation than low affinity
editors. Hormones and reproduction in fishes, amphibians and
receptors. This may be the case in green turtles. However, a more
reptiles. NY: Plenum Press. p 145–169.
thorough investigation will be essential to test this hypothesis.
Jessob T, Limpus C, Whitter J. 1999. Plasma steroid interactions during
Other reproductive steroids, such as T and progesterone (P)
high density green turtle nesting and associated disturbance. Gen
have been detected in other green turtles at Ras Al-Hadd
(Al-Habsi et al., 2006). This has also been reported in other green
Lance V, Owens DW, Callard IP. 1979. Radioimmunoassay of plasma
turtles from different geographical populations (Lance et al., ’79;
progesterone, testosterone, total estrogens and immunoreactive
Licht et al., ’79, ’80; Wibbels et al., ’92; Jessob et al., ’99).
gonadotropin in the nesting and non-nesting green sea turtle
Although the hormone values between different nesting
Chelonia mydas (L). Experientia 35:1119–1120.
phases are not significant, individual values of the same phase
Licht P. 1982. Endocrine patterns in the reproductive cycles of turtles.
showed some variations which may be related to stress caused by
the long delays in oviposition. This behavior is related to the
Licht P, Rainy W, Cliffton K. 1980. Serum gonadotropins and steroids
repeated nest collapses which keep the turtle seeking new sites
associated with breeding activities in the green turtle Chelonia
until it excavates a nest without collapse. Repeated changes in
mydas II. Mating and nesting in natural populations. Gen Comp
nest sites can be exhausting and stressful (Alkindi et al., 2003).
In addition, E2 levels in the nesting green turtles are so low
Licht PJ, Wood J, Owens D, Wood F. 1979. Serum gonadotropins and
that any change, such as stress or disturbance, can significantly
steroids associated with breeding activities in the green turtle,
change the E2 levels very rapidly and significantly.
Chelonia mydas, in Captive animals. Gen Comp Endocrinol 39:
Owens (’97) suggested a general model relative to hormone
dynamics in female sea turtles during the reproductive cycle.
Limpus CJ. 1995. Global overview of the status of marine turtles: a
During vitellogenesis and maturation of the follicles, there is a
1995 viewpoint. In: Bjorndal KA, editor. Biology and conservation of
significant rise in E2 concentration. However, during migration
sea turtles, Revised edition. Washington, DC: Smithsonian Institu-
and nesting, there is a significant decline in the E2 level, which
Miller JD. 1997. Reproduction in sea turtles. In: Lutz PL, Musick JA,
Rostal D, Owens D, Grumbles J, MacKenzie S, Amoss M.
editors. The biology of sea turtles. Boca Raton, FL: CRC Press. p 51–81.
1998. Seasonal reproductive cycle of the Kemp’s Ridley sea
Owens D. 1976. The endocrine control of reproduction and growth in
the green sea turtle, Chelonia mydas. West Arizona. Tucson, AR:
Wibbels T, Owens DW, Limpus CJ, Reeds PC, Amos Jr MS. 1990.
Owens D. 1997. Hormones in the life history of the sea turtles. In:
Seasonal changes in serum gonadal steroids associated with
Lutz P, Ausick J, editors. Biology of sea turtles. New York, NY: CRC
migration, mating and nesting in the loggerhead sea turtle (Caretta
caretta). Gen Comp Endocrinol 79:154–164.
Owens D, Ruiz GJ. 1980. Obtaining blood and cerebrospinal fluid from
Wibbels T, Owens DW, Licht P, Limpus CJ, Reeds PC, Amos Jr MS.
marine turtles. Herpetologica 36:17–20.
1992. Serum gonadotropins and gonadal steroids associated with
Porthe-Nibelle J, Lahlou B. 1978. Uptake and binding of glucocorti-
ovulation and egg production in sea turtles. Gen Comp Endocrinol
coids in fish tissues. J Endocrinol 78:407–416.
A Guide to Computing with Statistics with SPSS Release 10 for Windows: With Supplements for Releases 8and 9, Dennis Howitt, Duncan Cramer, Prentice Hall PTR, 2000, 0130326852, 9780130326850, . This text isdesigned for statistics courses taught in departments of Psychology. SPSS Guide to Computing for Windows10.0 walks students through the process of data analysis in straightforward, intuitive st
www.valueram.com Kingston Branded and ValueRAM Meeting your memory needs. Kingston Technology offers two memory productdiffer in support and testing services. The informationlines: Kingston® branded and ValueRAM™. While bothbelow clarifies these differences and will assist you inproduct lines are made with premium componentsand are carefully designed and manufactured, they System