Seasonal changes in testicular size, plasma testosterone concentration and body weight in captive flying foxes (Pteropus poliocephalus and P. scapulatus) M. A. McGuckin and A. W. Blackshaw Department of Physiology and Pharmacology, University of Queensland, St Lucia, QUI 4067, Australia
Summary. Adult male flying foxes Pteropus poliocephalus and P. scapulatus were captured in south-east Queensland and kept in outdoor enclosures. Testicular size (TS), plasma testosterone concentrations (PTC) and body weight (BW) were measured over 1-year periods. Testicular recrudescence in P. poliocephalus began before the summer solstice and TS was greatest during mid-March (autumn) and lowest from July to September. Large increases in PTC were observed in all indiviuals \m=~\1 month after the peak in TS. BW also increased around the time of the mating season, changes being correlated significantly with changes in TS. Mating occurred between April and June, and births from late October to late November. In P. scapulatus, TS was greatest in the spring (October) and least in the autumn (February to May); PTC fluctuated throughout the year in this species but, unlike P. poliocephalus, did not show a single large increase in the mating season. BW showed a similar seasonal pattern to that seen in P. poliocephalus, being greatest at the time of greatest TS. Mating occurred in October to November, and births in autumn. In captivity, in outdoor enclosures, these species maintained the seasonal reproductive patterns observed in the wild. The 2 species respond differently to the same environmental cues in terms of regulation of the timing of their breeding seasons. Keywords: season; reproduction; flying fox; testis; testosterone
Introduction The megachiropterans known as flying foxes or fruit bats (genus Pteropus) are represented in Australia by 4 species: P. alecto, P. conspicillatus, P. poliocephalus and P. scapulatus. Flying foxes are gregarious animals, living together in large groups (known as camps) and showing complex social behaviour (Ratcliffe, 1931, 1932; Nelson, 1963; R. Puddicombe, 1981, unpublished). The grey-headed flying fox, P. poliocephalus, inhabits the coastal regions of eastern Australia from the Bass Strait Islands to around the Tropic of Capricorn on the Queensland coast (Hall, 1981). The smaller, little red flying fox, P. scapulatus, is found in the coastal and inland regions of eastern and northern Australia (Hall, 1981). These species are important pollinators of native forests (McWilliam, 1986) as well as being pests to commercial fruit orchards (Loebel & Sanewski, 1987). A greater understanding of their reproductive biology is essential for the development of future
management strategies. There have been few detailed studies of reproduction in these species. In Australia, grey-headed foxes show a synchronized breeding period throughout their range. Mating occurs in the
flying
*Present address: Department of Obstetrics and Gynaecology, Royal Brisbane Hospital. Herston, Qld 4006. Australia.
University of Queensland.
Clinical Sciences
Building,
(March and April) and births in late September and October (Ratcliffe, 1931; Nelson, 1963; Puddicombe, 1981; Towers & Martin, 1986). In the late summer, males and females gather in camps in preparation for mating. Adult males then spend considerable time 'acquiring' females for their harems; this involves much interaction with females and other males (Ratcliffe, 1931; Nelson, 1963; Puddicombe, 1981). During the mating period, males copulate frequently with the females in their harem within the camps during the day; some mating also occurs at feeding sites during the nightly forage. Testicular size alters seasonally in this species, being greatest during February and March and least from June to September (McGuckin & Blackshaw, 1987; Nelson, 1963). The breeding seasons of P. alecto and P. conspicillatus appear, on limited evidence, to have a similar timing to that of P. poliocephalus. In relation to the other 3 species, the breeding season in P. scapulatus is out of phase by ~6 months. Mating has been observed during October (Ratcliffe, 1931), in November and December (Nelson, 1963) and in January (Puddicombe, 1981); births have been observed during April and May (Nelson, 1963). Nelson (1963) found the greatest testicular size in December although his data did not include any samples collected from May to November, inclusive. Most of the information on reproduction in megachiropterans is derived from field studies of wild bat populations. Little information has been obtained from captive animals because of problems of establishing and maintaining reproductively active, healthy, captive colonies of these species. Serial collection of data from captive individuals provides much information that is difficult to extract from field studies. The following studies examined reproduction in male Pteropus maintained in captivity, in conditions as close as possible to those occurring naturally. autumn
Materials and Methods P. poliocephalus Data were collected from 7 adult males (identified as p21, 22, 23, 24, 25, 27 and 28) captured in south-east Queensland and kept in captivity for more than 1 year before the first sample. The bats were individually identified by numbered, stainless steel tags placed on a thumb claw. Animals were kept at a Brisbane farm and housed with adult females in an outdoor cage that provided areas of shelter and areas completely open to the environment. Their diet consisted of a daily supply of various diced fresh fruits (chiefly canteloupes. bananas and apples) supplemented with either Complan (Glaxo Australia) or milk powder. At each sampling time (around midday), animals were captured, a 2-ml blood sample was obtained from a vein in the flap of skin adjacent to the hind limb, testis dimensions were measured with Vernier calipers, always by the same person, and body weight was determined (to the nearest gram) using an electronic balance. Samples were obtained approximately once a month from July 1985 until June 1986; during March, April and May, samples were obtained at fortnightly intervals. Testicular volume was calculated using the formula for a prolate spheroid. Plasma was obtained by centrifugation at 4000 # for 15 min and kept at 20°C until thawed for assay. Plasma testosterone concentrations were determined by radioimmunoassay. Intra-assay coefficients of variation were 7-3 and 101% and interassay coefficients of variation were 19-9 and 171% (« 10) at the low and high levels of the standard curve, respectively. Vaginal smears were obtained from 10 females housed with these males on several occasions throughout the breeding season and were assessed for the presence of spermatozoa and seminal vesicle secretions. In both species the active seminal vesicles produce masses of globular material that are readily identified in vaginal smears (McGuckin, 1989). One individual (p23) was monitorecf from August 1984 until June 1987 as part of a preliminary investigation, during this experiment and during a later experiment. —
=
P.
scapulatus
scapulatus captured in central Queensland during April 1986 were kept in captivity (S13, 17, 23, a large group (~ 100 individuals including females) within a single outdoor cage and allowed to adapt to captivity before handling commenced. The diet for this species was similar to that provided for P. poliocephalus. Samples were obtained about every 20 days from August 1986 until April 1987 and then less frequently until August 1987. At each sampling time (around midday), animals were captured, a 1-2 ml blood sample was obtained, testis dimensions were measured and body weight was determined. Plasma testosterone concentrations were determined by radioimmunoassay, intra- and interassay coefficients of variation were 8-5 and 12-4% (n 6), respect¬ ively, at the low level of the standard curve. Vaginal smears were obtained from females housed with these males on Six adult male P.
34, 35 and 37) and housed in
=
several occasions vesicle secretions.
Testosterone
throughout
the
breeding
season
and
were
assessed for the presence of spermatozoa and seminal
radioimmunoassay
Radioimmunoassay of testosterone in the plasma of Pteropus has been reported (McGuckin & Blackshaw, 1987). The testosterone antiserum (6050) was prepared by R. I. Cox (CSIRO, Division of Animal Production, Prospect, NSW, Australia), who reported significant cross-reactions with only 5a-dihydrotestosterone (31%) and 4-androsten-3ß,17ß-diol (30%). Extraction and assay were performed in 12 75 mm glass tubes. Tritiated testosterone (103 Ci/mmol, Amersham Australia) was dissolved in assay buffer (01 M phosphate gelatin buffer, pH 7-4) to 10000c.p.m.,'100 pi. Aliquants (50 pi) of unknown plasma samples in triplicate were extracted in 2 ml diethyl ether.
Testosterone standards 0^1024 pg/tube in ethanol were dried under a stream of N2 in triplicate for each assay. Dried plasma extracts and standards were resuspended in 300 pi of antiserum, at a 1:20 000 dilution in assay buffer, and 100 µ of tracer. Tubes were incubated for 16 h at 4"C followed by addition of 400 pi of a dextran-coated charcoal suspension for 10 min at 4C. The supernatant was decanted, added to 3 ml of scintillation fluid (5-6% (v/v) acetic acid in toluene containing 0-3% (w/v) 2,5 diphenyloxazole and 003% (w/v) l,5-bis[2-(5-phenyloxazole)]-benzene, Syndel Laboratories, Canada) and radioactivity was measured in an LKB liquid scintillation counter. A standard curve was calculated using a logit-log transformation and testosterone concentration in plasma samples was determined after allowing for recovery. Plasma samples found to contain > 800 pg testosterone/50 pi were diluted in assay buffer and re-extracted and assayed. Sensitivity of 52 consecutive testosterone assays was 6-2 + 0-6 pg/tube, which is equivalent to 013 ng/ml of plasma when 50 µ plasma was assayed and extraction recovery was 90%. Charcoal-stripped castrate Pteropus plasma did not cause positive or negative blanking in the testosterone assay, and it was therefore not considered necessary to include stripped plasma as blanks in each assay and the zero standard was used as the assay blank. Recovery of testosterone was assessed by the addition of 256 pg to a plasma sample containing ~ 1 ng testosterone/ml before extraction in 5 assays. Recovery of added steroid was determined after allowing for extraction recovery and was 104 + 7 (s.d.)%. Parallelism was demonstrated using serial dilutions of a plasma sample containing ~10ng testosterone/ml. Celite chromatography (Thorneycroft et al, 1973) showed that the peaks in immuno¬ reactivity and radioactivity for testosterone corresponded. Adult male human, horse, cat, rat, dog and pig plasma was assayed; concentrations obtained were within the ranges reported for these species.
Statistics
Changes in the parameters measured over time were assessed for statistical significance using paired
; tests.
Results P. poliocephalus
Testicular volume changed with season, and mean testicular size was greatest during mid-March (significantly greater than at all other times, < 005, Fig. 1). The timing of greatest testicular size in individuals ranged from mid-March to mid-April although some individuals achieved close to their greatest testes as early as January. Testes were smallest during July-September, then gradually < 005) typically increasing in size more increased (greater in November than in September, after solstice. the summer rapidly Plasma testosterone concentrations were lowest from July to November and had returned to these concentrations by June of the following year. Mean concentrations increased slowly from December (December > November, < 005) and all < 001; February > January, concentrations in March Mean characterized increases were by during and/or April. large profiles mid-April were greater than those in late March and mid-May ( < 005). Body weight also changed considerably with season, being greatest around the time of mating. Body weight increased gradually from November and was greatest in March and April (P < 005); after April, it decreased rapidly until mid-May, reaching a minimum in July (P < 005). Mean body weight increased by 27% from July to mid-April. Body weights were correlated with testicular volumes both within individuals throughout the year and for the entire group. Within individuals, these correlations were highly significant (r 0-79, 0-73, 0-87, 0-89, 0-65, 0-94 and 0-90 for p21, 22, 23, 24, 25, 27 and 28, respectively; < 001) as was the correlation for the group (r 0-58, =
=
Summary. Adult male flying foxes Pteropus poliocephalus and P. scapulatus were captured in south-east Queensland and kept in outdoor enclosures. Testicular size (TS), plasma testosterone concentrations (PTC) and body weight (BW) were measured over 1-year periods. Testicular recrudescence in P. poliocephalus began before the summer solstice and TS was greatest during mid-March (autumn) and lowest from July to September. Large increases in PTC were observed in all indiviuals \m=~\1 month after the peak in TS. BW also increased around the time of the mating season, changes being correlated significantly with changes in TS. Mating occurred between April and June, and births from late October to late November. In P. scapulatus, TS was greatest in the spring (October) and least in the autumn (February to May); PTC fluctuated throughout the year in this species but, unlike P. poliocephalus, did not show a single large increase in the mating season. BW showed a similar seasonal pattern to that seen in P. poliocephalus, being greatest at the time of greatest TS. Mating occurred in October to November, and births in autumn. In captivity, in outdoor enclosures, these species maintained the seasonal reproductive patterns observed in the wild. The 2 species respond differently to the same environmental cues in terms of regulation of the timing of their breeding seasons. Keywords: season; reproduction; flying fox; testis; testosterone
Introduction The megachiropterans known as flying foxes or fruit bats (genus Pteropus) are represented in Australia by 4 species: P. alecto, P. conspicillatus, P. poliocephalus and P. scapulatus. Flying foxes are gregarious animals, living together in large groups (known as camps) and showing complex social behaviour (Ratcliffe, 1931, 1932; Nelson, 1963; R. Puddicombe, 1981, unpublished). The grey-headed flying fox, P. poliocephalus, inhabits the coastal regions of eastern Australia from the Bass Strait Islands to around the Tropic of Capricorn on the Queensland coast (Hall, 1981). The smaller, little red flying fox, P. scapulatus, is found in the coastal and inland regions of eastern and northern Australia (Hall, 1981). These species are important pollinators of native forests (McWilliam, 1986) as well as being pests to commercial fruit orchards (Loebel & Sanewski, 1987). A greater understanding of their reproductive biology is essential for the development of future
management strategies. There have been few detailed studies of reproduction in these species. In Australia, grey-headed foxes show a synchronized breeding period throughout their range. Mating occurs in the
flying
*Present address: Department of Obstetrics and Gynaecology, Royal Brisbane Hospital. Herston, Qld 4006. Australia.
University of Queensland.
Clinical Sciences
Building,
(March and April) and births in late September and October (Ratcliffe, 1931; Nelson, 1963; Puddicombe, 1981; Towers & Martin, 1986). In the late summer, males and females gather in camps in preparation for mating. Adult males then spend considerable time 'acquiring' females for their harems; this involves much interaction with females and other males (Ratcliffe, 1931; Nelson, 1963; Puddicombe, 1981). During the mating period, males copulate frequently with the females in their harem within the camps during the day; some mating also occurs at feeding sites during the nightly forage. Testicular size alters seasonally in this species, being greatest during February and March and least from June to September (McGuckin & Blackshaw, 1987; Nelson, 1963). The breeding seasons of P. alecto and P. conspicillatus appear, on limited evidence, to have a similar timing to that of P. poliocephalus. In relation to the other 3 species, the breeding season in P. scapulatus is out of phase by ~6 months. Mating has been observed during October (Ratcliffe, 1931), in November and December (Nelson, 1963) and in January (Puddicombe, 1981); births have been observed during April and May (Nelson, 1963). Nelson (1963) found the greatest testicular size in December although his data did not include any samples collected from May to November, inclusive. Most of the information on reproduction in megachiropterans is derived from field studies of wild bat populations. Little information has been obtained from captive animals because of problems of establishing and maintaining reproductively active, healthy, captive colonies of these species. Serial collection of data from captive individuals provides much information that is difficult to extract from field studies. The following studies examined reproduction in male Pteropus maintained in captivity, in conditions as close as possible to those occurring naturally. autumn
Materials and Methods P. poliocephalus Data were collected from 7 adult males (identified as p21, 22, 23, 24, 25, 27 and 28) captured in south-east Queensland and kept in captivity for more than 1 year before the first sample. The bats were individually identified by numbered, stainless steel tags placed on a thumb claw. Animals were kept at a Brisbane farm and housed with adult females in an outdoor cage that provided areas of shelter and areas completely open to the environment. Their diet consisted of a daily supply of various diced fresh fruits (chiefly canteloupes. bananas and apples) supplemented with either Complan (Glaxo Australia) or milk powder. At each sampling time (around midday), animals were captured, a 2-ml blood sample was obtained from a vein in the flap of skin adjacent to the hind limb, testis dimensions were measured with Vernier calipers, always by the same person, and body weight was determined (to the nearest gram) using an electronic balance. Samples were obtained approximately once a month from July 1985 until June 1986; during March, April and May, samples were obtained at fortnightly intervals. Testicular volume was calculated using the formula for a prolate spheroid. Plasma was obtained by centrifugation at 4000 # for 15 min and kept at 20°C until thawed for assay. Plasma testosterone concentrations were determined by radioimmunoassay. Intra-assay coefficients of variation were 7-3 and 101% and interassay coefficients of variation were 19-9 and 171% (« 10) at the low and high levels of the standard curve, respectively. Vaginal smears were obtained from 10 females housed with these males on several occasions throughout the breeding season and were assessed for the presence of spermatozoa and seminal vesicle secretions. In both species the active seminal vesicles produce masses of globular material that are readily identified in vaginal smears (McGuckin, 1989). One individual (p23) was monitorecf from August 1984 until June 1987 as part of a preliminary investigation, during this experiment and during a later experiment. —
=
P.
scapulatus
scapulatus captured in central Queensland during April 1986 were kept in captivity (S13, 17, 23, a large group (~ 100 individuals including females) within a single outdoor cage and allowed to adapt to captivity before handling commenced. The diet for this species was similar to that provided for P. poliocephalus. Samples were obtained about every 20 days from August 1986 until April 1987 and then less frequently until August 1987. At each sampling time (around midday), animals were captured, a 1-2 ml blood sample was obtained, testis dimensions were measured and body weight was determined. Plasma testosterone concentrations were determined by radioimmunoassay, intra- and interassay coefficients of variation were 8-5 and 12-4% (n 6), respect¬ ively, at the low level of the standard curve. Vaginal smears were obtained from females housed with these males on Six adult male P.
34, 35 and 37) and housed in
=
several occasions vesicle secretions.
Testosterone
throughout
the
breeding
season
and
were
assessed for the presence of spermatozoa and seminal
radioimmunoassay
Radioimmunoassay of testosterone in the plasma of Pteropus has been reported (McGuckin & Blackshaw, 1987). The testosterone antiserum (6050) was prepared by R. I. Cox (CSIRO, Division of Animal Production, Prospect, NSW, Australia), who reported significant cross-reactions with only 5a-dihydrotestosterone (31%) and 4-androsten-3ß,17ß-diol (30%). Extraction and assay were performed in 12 75 mm glass tubes. Tritiated testosterone (103 Ci/mmol, Amersham Australia) was dissolved in assay buffer (01 M phosphate gelatin buffer, pH 7-4) to 10000c.p.m.,'100 pi. Aliquants (50 pi) of unknown plasma samples in triplicate were extracted in 2 ml diethyl ether.
Testosterone standards 0^1024 pg/tube in ethanol were dried under a stream of N2 in triplicate for each assay. Dried plasma extracts and standards were resuspended in 300 pi of antiserum, at a 1:20 000 dilution in assay buffer, and 100 µ of tracer. Tubes were incubated for 16 h at 4"C followed by addition of 400 pi of a dextran-coated charcoal suspension for 10 min at 4C. The supernatant was decanted, added to 3 ml of scintillation fluid (5-6% (v/v) acetic acid in toluene containing 0-3% (w/v) 2,5 diphenyloxazole and 003% (w/v) l,5-bis[2-(5-phenyloxazole)]-benzene, Syndel Laboratories, Canada) and radioactivity was measured in an LKB liquid scintillation counter. A standard curve was calculated using a logit-log transformation and testosterone concentration in plasma samples was determined after allowing for recovery. Plasma samples found to contain > 800 pg testosterone/50 pi were diluted in assay buffer and re-extracted and assayed. Sensitivity of 52 consecutive testosterone assays was 6-2 + 0-6 pg/tube, which is equivalent to 013 ng/ml of plasma when 50 µ plasma was assayed and extraction recovery was 90%. Charcoal-stripped castrate Pteropus plasma did not cause positive or negative blanking in the testosterone assay, and it was therefore not considered necessary to include stripped plasma as blanks in each assay and the zero standard was used as the assay blank. Recovery of testosterone was assessed by the addition of 256 pg to a plasma sample containing ~ 1 ng testosterone/ml before extraction in 5 assays. Recovery of added steroid was determined after allowing for extraction recovery and was 104 + 7 (s.d.)%. Parallelism was demonstrated using serial dilutions of a plasma sample containing ~10ng testosterone/ml. Celite chromatography (Thorneycroft et al, 1973) showed that the peaks in immuno¬ reactivity and radioactivity for testosterone corresponded. Adult male human, horse, cat, rat, dog and pig plasma was assayed; concentrations obtained were within the ranges reported for these species.
Statistics
Changes in the parameters measured over time were assessed for statistical significance using paired
; tests.
Results P. poliocephalus
Testicular volume changed with season, and mean testicular size was greatest during mid-March (significantly greater than at all other times, < 005, Fig. 1). The timing of greatest testicular size in individuals ranged from mid-March to mid-April although some individuals achieved close to their greatest testes as early as January. Testes were smallest during July-September, then gradually < 005) typically increasing in size more increased (greater in November than in September, after solstice. the summer rapidly Plasma testosterone concentrations were lowest from July to November and had returned to these concentrations by June of the following year. Mean concentrations increased slowly from December (December > November, < 005) and all < 001; February > January, concentrations in March Mean characterized increases were by during and/or April. large profiles mid-April were greater than those in late March and mid-May ( < 005). Body weight also changed considerably with season, being greatest around the time of mating. Body weight increased gradually from November and was greatest in March and April (P < 005); after April, it decreased rapidly until mid-May, reaching a minimum in July (P < 005). Mean body weight increased by 27% from July to mid-April. Body weights were correlated with testicular volumes both within individuals throughout the year and for the entire group. Within individuals, these correlations were highly significant (r 0-79, 0-73, 0-87, 0-89, 0-65, 0-94 and 0-90 for p21, 22, 23, 24, 25, 27 and 28, respectively; < 001) as was the correlation for the group (r 0-58, =
=
