Mice with either diminished or elevated levels of anti-Müllerian hormone have decreased litter sizes

Mice with either diminished or elevated levels of anti-Müllerian hormone have decreased litter... Abstract Anti-Müllerian hormone (AMH) is both a gonadal hormone and a putative paracrine regulator of neurons, the uterus, and the placenta. A mouse line with neuronal expression of AMH (Thy1.2-AMH) was generated to examine the role of paracrine AMH in the brain. The mice had normal behavior, but unexpectantly AMH was present in the circulation of the transgenic mice. Thy1.2-AMHTg/0 studs sired pups with a normal frequency, when mated with wild-type dams. In stark contrast, Thy1.2-AMHTg/0 dams rarely gave birth, with evidence of spontaneous midgestational abortion. This leads to the hypothesis that AMH influences the capacity of dams to carry concepti to term. This hypothesis was tested by mating AMH-deficient (Amh−/−), Thy1.2-AMHTg/0, and wild-type dams when 49-, 80-, and 111 days old, using proven wild-type studs. The litter sizes from the first two matings and the number of fetuses present on the 10th day of gestation of the third mating were recorded. Thy1.2-AMHTg/0 dams carried near normal numbers of midterm fetuses, but typically produced no pups, indicating that extensive late resorption of fetuses was occurring. Amh−/− dams exhibited a lesser reduction in litter size than the Thy1.2-AMHTg/0 dams, with no evidence of enhanced loss of fetuses. In conclusion, this study provides the first evidence that high AMH levels can cause a miscarriage phenotype and that the absence of AMH affects reproductive output. Introduction Anti-Müllerian hormone (AMH) is a sexually dimorphic gonadal hormone [1, 2], which is also synthesized by various nongonadal cells, including some mature neurons [3–5], the uterus [6], and the placenta [7]. Circulating AMH is entirely derived from the gonads [8, 9], suggesting that the nongonadal sources of AMH act exclusively as paracrine regulators. In mature females, both the endocrine and paracrine sources of AMH are linked to reproduction. Circulating ovarian AMH exhibits minimal variation during the ovarian cycle [10], but declines acutely during the second and third trimester of pregnancy [11–13], and chronically with age, in parallel with the decline in the pool of growing ovarian follicles [14]. AMH slows reproductive senescence in mice [15], and modulates the influence of follicle-stimulating hormone (FSH) on granulosa cells [16]. However, the physiological actions of AMH may extend to other reproductive organs, as the AMH-specific receptor (AMHR2) is expressed in the mammary gland [17], uterus [6, 18], and placenta [7]. Speculatively, the age- and/or pregnancy-related decline in circulating AMH may modulate the transfer of resources from the mother to her offspring. Equally, the uterus and placenta synthesize AMH [6, 7], raising the possibility that the uterine and placental receptors are predominantly or exclusively activated by locally produced (paracrine) AMH. Transgenic mice with neuron-specific expression of human AMH (hAMH) were generated to study the role of paracrine AMH in the mature brain. The receptor-binding region of hAMH is 93% homologous to the murine protein and activates murine AMH receptors in multiple experimental systems [19–21]. The Thy1.2 promoter was used to delay the expression of the transgene until neurons had matured [22], thus mimicking the normal developmental onset of neuronal AMH expression, and avoiding the virilizing effect of AMH on the developing nervous system [23–25]. We report here that the Thy1.2-AMHTg/0mice contain transgenic hAMH in their circulation and exhibit a high rate of midterm loss of embryos. Materials and methods Thy1.2-AMH mice A cDNA clone containing the full-length Homo sapiens AMH mRNA was purchased from InVitrogen/Life Technologies (MGC:54308; IMAGE:5168299; Unigene Cluster ID: Hs.112432; Carlsbad, CA, USA). The AMH gene was cloned into the mouse thymus antigen 1.2 (Thy1.2) expression cassette [22] and the Thy1.2-AMH construct verified by sequencing (The Allan Wilson Centre, Massey University, Palmerston North, New Zealand). Transgenic mouse lines were then produced by microinjection of zygotes from FVB mice with the Thy1.2-AMH cDNA by AgResearch Ltd, Ruakura, Hamilton, New Zealand. The Thy1.2-AMH mouse lines were maintained by cross-breeding with Amh+/+ mice, which were the wild-type offspring from Amh+/-x Amh+/− matings, in order to generate congenic Amh and Thy1.2-AMH lines. The resulting pups were therefore either hemizygous for the transgene (Thy1.2-AMHTg/0) or wild type (Thy1.2-AMH0/0). The mice were genotyped by PCR using a forward primer located in the Thy1.2 promoter (5’-TGT AGC TTT CCC CAC CAC AGA-3΄) and a reverse primer in the AMH transgene (5’-AGG GCC TCA GTC CCC AGC AGA-3΄), with the amplified sequence being 145 bp. The Amh−/− mouse line, which carries a targeted gene disruption that globally abolishes AMH expression [26], was obtained from The Jackson Laboratory (B6;129S7-Amhtm1Bhr/J, Stock number 002187, Bar Harbor, ME, USA). All mice were housed in M.I.C.E. cages (Animal Care Systems, CO 80112, USA), as previously described [27]. The temperature was 21 ± 1°C and the relative humidity varied between 40 and 50%. The housing met all requirements specified in the National Research Council's Guide for the Care and Use of Laboratory Animals (NIH, USA) and in the guidelines of the New Zealand Ministry for Primary Industries. The experiments were approved by the Animal Ethics Committee of the University of Otago. Human anti-Müllerian hormone in the blood of Thy1.2-AMH mice The mice were terminally anaesthetized with ketamine (225 mg/kg, Phoenix Pharm) and Domitor (3 mg/kg, Medetomidine, Pfizer Animal Health), after which blood was drawn by cardiac puncture and clotted at room temperature for 1 h. The serum was aliquoted and snap-frozen in liquid nitrogen for storage at –80°C. The transgenic hAMH levels were assayed using a method that does not detect endogenous murine AMH [28]. Briefly, the samples were diluted 1:24 and were assayed using the PicoAMH ELISA (Cat# AL-124-I; Ansh Labs) as per the manufacturer's procedure, with the exception that the antibody-biotin conjugate was replaced with the antibody-biotin conjugate (100 μL) from the AMH Gen II ELISA kit (Cat# A79765; BeckmanCoulter). The picoAMH ELISA capture antibody produces high sensitivity and the Gen II detection antibody circumvents the variable ability of the picoAMH detection antibody to recognize different recombinant hAMH preparations. Quantitative reverse transcriptase-PCR detection of human anti-Müllerian hormone expression Tissue specificity The adrenal glands, brain, heart, kidney, liver, lung, skeletal muscle, and the uterus from four Thy1.2-AMHTg/0 were rapidly dissected, snap-frozen in liquid nitrogen, after which the mRNA was extracted and cDNAs generated as previously described [29]. The primers were designed to amplify the hAMH transgene but not murine Amh cDNA, with cDNA from the brains of wild-type mice used to verify primer specificity (F 5΄ AGGAAGTGACCTGGGAGCAACA 3΄, R 5΄ AGCCCAGCCCTCGTCACAGTGA 3΄). Hypoxanthine-guanine phosphoribosyltransferase (Hprt) expression (F 5΄ CTGGTGAAAAGGACC-TCTCG 3΄, R 5΄ TGAAGTACTCATTATAGTCAAGGGCA 3΄) was used to normalize data for mRNA abundance. Temporal specificity A portion of the brains from 0-, 5-, 15-, 22-, 30-, 42-, and 53-day-old Thy1.2-AMHTg/0 mice were processed as described in the previous paragraph, with the exception that glyceraldehyde 3-phosphate deydrogenase (Gapdh) expression (F 5΄ TCTTCACCACCATGGAGAAG 3΄, 5΄ ACCAAAGTTGTCATGGAT-GAC 3΄) was used to normalize the data for mRNA abundance. Localization of human anti-Müllerian hormone and endogenous anti-Müllerian hormone protein The localization of the transgenic hAMH protein was examined by immunohistochemistry using perfusion-fixed tissues, as previously described [25]. The primary antibody was a goat peptide antibody made to the N-terminal (pro region) of human MIS/AMH (#AF2748, R&D Systems, Minneapolis, MN, Supplementary Table 1). Murine endogenous AMH protein was detected with the goat polyclonal MIS-C20 antibody (Santa Cruz, Dallas, Tx) using Bouins-fixed ovaries from mice in estus. Nonimmune goat IgG and tissues from wild-type mice were used as negative controls. Observations were made using a Zeiss Axioplan microscope, equipped with 0.075, 0.75, and 1.40 numerical aperture plan-apochromat lens for bright-field observations and 0.30 and 0.50 numerical aperture neofluar lens for phase-contrast imaging. Photomicrographs were produced with a Zeiss AxioCam HRc RGB camera, with no digital correction. Breeding of the Thy1.2-AMH mice The Thy1.2-AMH mice were bred as stable stud and dam pairs, which were checked daily for births (pups or postpartum bleeding). The colony records were maintained in an in-house computer programme which includes the date of birth, date of mating, date of birth of each litter, litter size, date and cause of death. The programme uses these data to track lineages and multiple parameters relating to breeding performance, including time between mating and birth, interval between births and litter size. The first analysis of the reproduction of the Thy1.2-AMH colony was from these records, and corresponds to the period when the genetic background was being backcrossed from FVB to that of the Amh colony. Mating experiment Six Thy1.2-AMHTg/0, Thy1.2-AMH0/0 (wild-type), and Amh+/+ (wild-type) dams and 12 Amh−/− dams were paired with a proven wild-type stud for mating over a 9-day period. The dams were transferred to the stud's cage, with each stud only having one partner at any one time. Eleven wild-type studs were used (AMH+/+ mice or Thy1.2-AMH0/0 mice, derived from either of the congenic colonies), and were mated at random with the dams. The dams were mated when aged 49-, 80- and 111 days old. All pups derived from the first two matings were culled at birth, and the litter sizes recorded. During the third mating, the dams were examined daily for the presence of a postcoital plug, and euthanized 10 days after the detection of the plug. The number of fetuses being carried was recorded. Statistical analysis The difference in the frequency of births within 50 days of mating (Figure 4) were examined using a Binomial Test, with an expected frequency of 1 (frequency observed in Amh+/+ dams). The influence of genotype on the frequency of dams with no pups or fetuses was examined using a Pearson's Chi squared test, with pair-wise comparison using a Fischer exact test when a statistically significant effect of genotype was observed (P < 0.05) (Figure 5). The differences in the sizes of litters generated by Thy1.2-AMHTg/0 studs and dams were examined using a one-way ANOVA. When three groups of dams were compared, a two-way ANOVA was used: pair-wise comparisons of genotype were made using one-way ANOVA when the two-way ANOVA was significant for genotype. The litter size data were also analyzed using a three-way ANOVA approach to take parity into account; the three factors were genotype, nulliparous (yes/no), and offspring stage (i.e., pup count for litter 1 and 2 or E10 embryo for litter 3). IBM SPSS Statistics v21 and Prism 7 (GraphPad Software Inc) were used for the statistical calculations. Results Only two of the 25 founder mice contained the transgene: both were female. One founder failed to give birth after 139 days of mating, despite the stud being replaced after 60 days. The other founder gave birth to a litter of seven pups 34 days after being paired with a stud, with a single pup being born 25 days later from a second pregnancy. The dam was culled at this stage, and a colony established from the Thy1.2-AMHTg/0 mice born from the first litter. The Thy1.2-AMHTg/0 mice exhibited no gross phenotype, with the reproductive tract of female Thy1.2-AMHTg/0 being of normal size and anatomy. The Thy1.2-AMHTg/0 mice were of normal appearance and weight, and exhibited no overt neurological symptoms or atypical behaviors. Their behavior was quantitatively normal in a range of tests, including open-field, rotarod, elevated plus maze, and forced swim test, as was the response of the mice to selected neuroactive drugs (Supplemental Information File 1). The expression of hAMH mRNA was first detected in the brains of 21-day-old Thy1.2-AMHTg/0 mice, with the level of expression increasing markedly between days 30 and 42 (Figure 1). Human AMH protein was expressed in most neurons of adult Thy1.2-AMHTg/0 mice, with the notable exception of the thalamus and hypothalamus (Figure 2; Supplemental Information Files 2 and 3). In the cerebral cortex, for example, the neurons in the multiform (VI) and internal pyramidal cell (V) layers were strongly stained by the anti-AMH antibody, as were most but not all of the external pyramidal cells. There was, however, minimal or no immunoreactivity in the internal (II) and external granular (IV) layers. The transgene was also detected in a minority of pituitary cells (Supplemental File 3). The mean levels of hAMH mRNA in brain were 8.5% of housekeeping gene (Hprt) mRNA levels but were less than 0.5% in adrenal glands, heart, kidney, liver, skeletal muscle, and uterus. Slightly higher levels were observed in the lung (0.9% of Hprt mRNA levels), which is consistent with prior reports of Thy1.2 promoter-driven transgene expression [22]. Unexpectedly, serum from adult (44–98 day old) Thy1.2-AMHTg/0 mice contained 1.12 ± 0.14 nmol/L (n = 14) of hAMH. The serum of the wild-type littermates contained undetectable levels of hAMH, as expected (<0.02 pM, n = 16). AMH in the blood is a mixture of uncleaved proAMH and AMHR2-receptor-competent AMHN,C: the molecular forms of the hAMH in the blood of theThy1.2-AMHTg/0mice mirrors that of AMH in human serum [28]. The level of endogenous AMH expression in the ovary was largely similar in Thy1.2-AMHTg/0 mice when compared to Thy1.2-AMH0/0 females by immunohistochemistry (Figure 3). At most stages of follicle development, similar levels of AMH expression were observed with the exception of primary follicles which show weak staining in wild-type females but no staining in Thy1.2-AMH0/0 mice. Figure 1. View largeDownload slide The expression of the hAMH transgene with postnatal age. The level of hAMH mRNA in the brains of Thy1.2-AMHTg/0 mice were measured by quantitative PCR, and normalized to the level of Gapdh mRNA. The values for individual mice are illustrated, using blue dots for males (M) and pink squares for females (F). The number of mice at each age group was day 0, 2F, 2M; day 5, 1F, 4M; day 15, 1F, 1M; day 22 2F, 1M; day 30 3F, 2M; day 42 2M; day 53 2F. Figure 1. View largeDownload slide The expression of the hAMH transgene with postnatal age. The level of hAMH mRNA in the brains of Thy1.2-AMHTg/0 mice were measured by quantitative PCR, and normalized to the level of Gapdh mRNA. The values for individual mice are illustrated, using blue dots for males (M) and pink squares for females (F). The number of mice at each age group was day 0, 2F, 2M; day 5, 1F, 4M; day 15, 1F, 1M; day 22 2F, 1M; day 30 3F, 2M; day 42 2M; day 53 2F. Figure 2. View large Download slide hAMHTg expression in the cerebral cortex. The hAMH was detected by immunohistochemistry utilizing an antibody to human proAMH. Strong staining was observed in the cortex of Thy1.2-AMHTg/0 mice (A) but little or no staining was observed in the cortex of wild-type (Thy1.2-AMH0/0) mice (B). The locations of the cortical layers are indicated by roman numerals; layer I: molecular layer; layer II: external granular layer; layer III: external pyramidal layer; layer IV: internal granular layer; layer V: internal pyramidal layer; layer VI: multiform layer and WM; white matter. Staining can be observed in the granular neuron cell bodies of layers I and IV, the pyramidal neurons in layer V and cell bodies at all depths of layer V. The density of neurons in the molecular layer is low but the occasional AMH-positive cell can be observed in layer I, which is consistent with neuronal staining. Little or no staining was observed in the pyramidal neurons of layer III. Higher magnification images of layer II (C), layer IV-V (D), and layer VI (E) reveals staining consistent with neuronal cell bodies including the initial segment of neuronal processes in the more intensely stained cells (arrowheads). No nonspecific staining of the brain was observed with goat antibody isotype controls (Supplemental file 2). The scale bars represent 100 μm. Figure 2. View large Download slide hAMHTg expression in the cerebral cortex. The hAMH was detected by immunohistochemistry utilizing an antibody to human proAMH. Strong staining was observed in the cortex of Thy1.2-AMHTg/0 mice (A) but little or no staining was observed in the cortex of wild-type (Thy1.2-AMH0/0) mice (B). The locations of the cortical layers are indicated by roman numerals; layer I: molecular layer; layer II: external granular layer; layer III: external pyramidal layer; layer IV: internal granular layer; layer V: internal pyramidal layer; layer VI: multiform layer and WM; white matter. Staining can be observed in the granular neuron cell bodies of layers I and IV, the pyramidal neurons in layer V and cell bodies at all depths of layer V. The density of neurons in the molecular layer is low but the occasional AMH-positive cell can be observed in layer I, which is consistent with neuronal staining. Little or no staining was observed in the pyramidal neurons of layer III. Higher magnification images of layer II (C), layer IV-V (D), and layer VI (E) reveals staining consistent with neuronal cell bodies including the initial segment of neuronal processes in the more intensely stained cells (arrowheads). No nonspecific staining of the brain was observed with goat antibody isotype controls (Supplemental file 2). The scale bars represent 100 μm. Figure 3. View largeDownload slide AMH immunohistochemical staining in Thy1.2-AMH0/0 (A, B, D, F) and Thy1.2-AMHTg/0 (C, E, D) mouse ovaries. The level of AMH immunostaining is similar when follicles are compared at the preantral stage (A, C), the small antral stage (D, E), and the medium-sized antral stage (F, G). Weak AMH immunoreactivity was present in the primary follicles of Thy1.2-AMH0/0 mice (B, arrowhead) but no immunostaining was observed at this stage in Thy1.2-AMHTg/0 mice (C, arrowhead). AMH was visualized with a diaminobenzidine chromogen (brown) with a hemotoxylin counterstain (blue). Images are representative of ovaries from three mice from each genotype, collected while in estrus. The scale bar represents 100 μm. Figure 3. View largeDownload slide AMH immunohistochemical staining in Thy1.2-AMH0/0 (A, B, D, F) and Thy1.2-AMHTg/0 (C, E, D) mouse ovaries. The level of AMH immunostaining is similar when follicles are compared at the preantral stage (A, C), the small antral stage (D, E), and the medium-sized antral stage (F, G). Weak AMH immunoreactivity was present in the primary follicles of Thy1.2-AMH0/0 mice (B, arrowhead) but no immunostaining was observed at this stage in Thy1.2-AMHTg/0 mice (C, arrowhead). AMH was visualized with a diaminobenzidine chromogen (brown) with a hemotoxylin counterstain (blue). Images are representative of ovaries from three mice from each genotype, collected while in estrus. The scale bar represents 100 μm. The Thy1.2-AMHTg/0 studs and dams were mated with Amh+/+ mice to change the genetic background of the Thy1.2-AMH colony. Breeding colony data were analyzed retrospectively and the breeding performance of the Thy1.2-AMHTg/0studs and dams was shown to be distinct. When Thy1.2-AMHTg/0 studs were mated with Amh+/+ dams, all dams gave birth within 50 days of entering the stud's cage, whereas only 5 of 18 Thy1.2-AMHTg/0 dams mated with Amh+/+ studs gave birth within this time frame (Figure 4A). The litter size was also markedly smaller when the transgene was carried by the dam rather than the stud (Figure 4B), independent of the number of back crosses. The frequency of Thy1.2-AMHTg/0 pups in litters that were genotyped was close to the Mendelian ratio (49.1%, n = 242), indicating that Thy1.2-AMHTg/0 fetuses were not selectively lost in utero. The time between the stud-dam pairing and birth was independent of which sex carried the transgene (Figure 4C). Two pregnant Thy1.2-AMHTg/0dams were euthanized when they failed to give birth. They were carrying one or three healthy fetuses that were atypically large. Postmortem examination of some of the barren dams revealed the presence of fetuses that appeared to have died around midgestation. Figure 4. View largeDownload slide Breeding statistics from the Thy1.2-AMH colony, with reference to whether the dam or the stud carried the hAMH transgene. (A) The percentage of pairings involving Thy1.2-AMHTg/0studs and Thy1.2-AMHTg/0 dams that yielded a litter within 50 days of pairing with a wild-type mouse. (B) The average litter size of Thy1.2-AMHTg/0 dams and wild-type dams mated with Thy1.2-AMHTg/0 studs. (C) The average time between mating and birth. Day 0 for the first litter was the first day of pairing, whereas day zero for the second litter was the day of birth of the first litter 1. The data are the mean plus the standard error of the mean, with the number of mice indicated on the bar. The 2nd+ pairing column contains data from the second litter onwards hence the number of data points exceeds the number of stud-dam pairings. ‘#’The frequency of births by the Thy1.2-AMHTg/0is significantly less than the observed frequency (1.0) of Amh+/+ dams, mated to Thy1.2-AMHTg/0 studs: # P < 0.0005 (Binomial Test). The litter size was significantly different depending on whether the transgene was carried by the stud or the dam: *First litter P = 0.002; **Second litter P < 0.005 (one-way ANOVA). There was no significant difference in the time to give birth (first litter P = 0.76, second litter P = 0.20). Figure 4. View largeDownload slide Breeding statistics from the Thy1.2-AMH colony, with reference to whether the dam or the stud carried the hAMH transgene. (A) The percentage of pairings involving Thy1.2-AMHTg/0studs and Thy1.2-AMHTg/0 dams that yielded a litter within 50 days of pairing with a wild-type mouse. (B) The average litter size of Thy1.2-AMHTg/0 dams and wild-type dams mated with Thy1.2-AMHTg/0 studs. (C) The average time between mating and birth. Day 0 for the first litter was the first day of pairing, whereas day zero for the second litter was the day of birth of the first litter 1. The data are the mean plus the standard error of the mean, with the number of mice indicated on the bar. The 2nd+ pairing column contains data from the second litter onwards hence the number of data points exceeds the number of stud-dam pairings. ‘#’The frequency of births by the Thy1.2-AMHTg/0is significantly less than the observed frequency (1.0) of Amh+/+ dams, mated to Thy1.2-AMHTg/0 studs: # P < 0.0005 (Binomial Test). The litter size was significantly different depending on whether the transgene was carried by the stud or the dam: *First litter P = 0.002; **Second litter P < 0.005 (one-way ANOVA). There was no significant difference in the time to give birth (first litter P = 0.76, second litter P = 0.20). On the basis of the breeding performance of the Thy1.2-AMHTg/0mice, AMH was postulated to regulate litter size, at least in part by influencing whether or not a conceptus is carried to term. This hypothesis was tested with a prospective experiment that compared the reproductive output of dams with AMH over expression (Thy1.2-AMHTg/0) with wild-type dams (Thy1.2-AMH0/0, Amh+/+) and dams with a null mutation in the Amh gene (Amh−/−). The dams were mated with a proven wild-type stud for 9 days, when aged 49, 80, and 111 days. The Thy1.2-AMH0/0 and Amh+/+dams were combined as a single wild-type group, as the colonies were congenic and as none of their reproductive traits were statistically different. The output from the Thy1.2-AMHTg/0 dams in this experiment (Figure 5) mirrored the results obtained from the colony record (Figure 4). Only one litter of two pups was obtained from two matings of the six Thy1.2-AMHTg/0 dams: the number of Thy1.2-AMHTg/0dams that did not give birth (five out of six) was statistically significant from both the wild-type and Amh−/−groups (P = 0.001), which all gave birth to at least one litter and were not significantly different to each other. On the third mating, all six female Thy1.2-AMHTg/0 dams exhibited copulatory plugs, indicating that mating had occurred and four out of six were carrying fetuses on day 10 of the third mating (Figure 5). Despite the severely reduced capacity of Thy1.2-AMHTg/0 dams to produce live offspring, there was no significant effect of genotype across the three groups in the frequency of pregnancy on embryonic day 10. Figure 5. View largeDownload slide Comparison of litter sizes from three groups of dams. Each dam was mated with a proven wild-type stud for 9 days on three occasions when aged 49-, 80- and 111 days old, with the data showing the number of pups born in each of the first two litters and the number of fetuses present on the 10h day of gestation of the third pregnancy. The blue dots represent individual wild-type dams (six Amh+/+ and six Thy1.2-AMH0/0): the two wild-type groups were congenic, and were not statistically different from each other. Zero data points represent dams that either did not give birth after one of the first two pairings with a stud or were not pregnant after the third mating. *There was a statistically significant effect of genotype on the frequency of dams with no pups from the first two matings (P < 0.0005 Pearson's Chi square test). In posthoc pairwise comparisons, the Thy1.2-AMHTg/0 dams were significantly different to the wild-type group (P = 0.003) and the Amh−/− group (P < 0.0005) (two-tailed Fischer exact test). The wild-type and Amh−/− groups were not significantly different (P = 0.27). There was no significant effect of genotype on dams with the presence of at least one fetus at E10 for the third mating (P = 0.56). Figure 5. View largeDownload slide Comparison of litter sizes from three groups of dams. Each dam was mated with a proven wild-type stud for 9 days on three occasions when aged 49-, 80- and 111 days old, with the data showing the number of pups born in each of the first two litters and the number of fetuses present on the 10h day of gestation of the third pregnancy. The blue dots represent individual wild-type dams (six Amh+/+ and six Thy1.2-AMH0/0): the two wild-type groups were congenic, and were not statistically different from each other. Zero data points represent dams that either did not give birth after one of the first two pairings with a stud or were not pregnant after the third mating. *There was a statistically significant effect of genotype on the frequency of dams with no pups from the first two matings (P < 0.0005 Pearson's Chi square test). In posthoc pairwise comparisons, the Thy1.2-AMHTg/0 dams were significantly different to the wild-type group (P = 0.003) and the Amh−/− group (P < 0.0005) (two-tailed Fischer exact test). The wild-type and Amh−/− groups were not significantly different (P = 0.27). There was no significant effect of genotype on dams with the presence of at least one fetus at E10 for the third mating (P = 0.56). When the matings that resulted in pups or fetuses were analyzed, the Amh−/− dams on average carried fewer offspring (pups or fetuses) than the wild-type dams, by between 1.1 and 1.4 offspring per mating. The effect of Amh genotype was strongly significant when the three matings were examined as a group (P = 0.001), after correction for the influence of parity and whether the dam was examined at E10 or at birth (Figure 6)1. The influence of Amh genotype was also strongly evident if the analysis was restricted to the first two matings, which were examined at birth (P = 0.002, Fig 6). The low frequency of detection of offspring precluded any statistical examination of the litter size of the Thy1.2-AMHTg/0dams. Figure 6. View largeDownload slide Influence of a null mutation of Amh on reproductive output by dams. The experiment is as described in Figure 5. Dams were only included in the analysis if birth occurred (pairing 1 and 2) or if fetuses were detected at E10 from the third mating. The data is the mean plus the standard error of the mean, with the number of mice indicated on the bar. *There was a significant effect of genotype (P = 0.001) in a three-way ANOVA, controlling for whether the dam was nulliparous and whether fetuses or pups were examined.1 # There was a significant effect of genotype (P = 0.002) and parity (P = 0.008), with no significant genotype-by-parity interaction (P = 0.65) (two-way ANOVA). When the wild-type and Amh−/− dams were compared separately for each mating, the P values were as follows: first P = 0.030; second P = 0.035, third P = 0.058 (one-way ANOVA). Figure 6. View largeDownload slide Influence of a null mutation of Amh on reproductive output by dams. The experiment is as described in Figure 5. Dams were only included in the analysis if birth occurred (pairing 1 and 2) or if fetuses were detected at E10 from the third mating. The data is the mean plus the standard error of the mean, with the number of mice indicated on the bar. *There was a significant effect of genotype (P = 0.001) in a three-way ANOVA, controlling for whether the dam was nulliparous and whether fetuses or pups were examined.1 # There was a significant effect of genotype (P = 0.002) and parity (P = 0.008), with no significant genotype-by-parity interaction (P = 0.65) (two-way ANOVA). When the wild-type and Amh−/− dams were compared separately for each mating, the P values were as follows: first P = 0.030; second P = 0.035, third P = 0.058 (one-way ANOVA). Discussion The regulation of litter size is complex, both with respect to cause and mechanism. Nulliparous and older dams tend to have smaller litters, but with variation between dams, even in inbred colonies [30, 31]. Environmental influences that alter energy balance also have large effects on litter size, which can involve trade-offs between the number and size of pups [31, 33]. Mechanistically, litter size is determined by multiple factors, including the number of oocytes released, the rate of implantation, and by resorption of concepti during either the embryonic or fetal stages [30, 32]. AMH influences the rate of depletion of ovarian follicles [15], which putatively influences when a female experiences age-related difficulty to conceive. The reproductive output of the Thy1.2-AMHTg/0 and Amh−/− dams raises the possibility that AMH may also be part of the molecular mechanism(s) that enable dams to adapt their reproductive output to their current circumstances. Reproductive phenotype of Thy1.2-AMHTg/0mice The Thy1.2-AMHTg/0dams, which overexpress AMH, infrequently gave birth and when birth occurred, the litter size was atypically small, in large part due to the resorption of fetuses. The loss of fetuses was not due to an inherent lack of viability of Thy1.2-AMHTg/0 fetuses, for two reasons. First, wild-type dams successfully carried Thy1.2-AMHTg/0 fetuses sired by Thy1.2-AMHTg/0 studs to term and second, pregnant Thy1.2-AMHTg/0 dams rarely carried their wild-type fetuses to term. Ovarian overexpression of AMH causes infertility, with restriction of primordial follicle activation being the putative cause [20]. The current data do not exclude the possibility that reduced follicle activation contributes to the phenotype of Thy1.2-AMHTg/0 dams, but pregnant Thy1.2-AMHTg/0dams carried too many fetuses at midgestation for it to be a major mechanism. The influence of the transgenic AMH in the current study was of similar magnitude in the nulliparous and multiparous dams, suggesting that AMH is not an essential part of the mechanism that diminishes the size of litters of nulliparous dams. Litter size of Amh−mh dams Mice that lacked the Amh gene also had smaller litters. The magnitude of the effect was, however, smaller than that observed in the Thy1.2-AMHTg/0dams, and there was no overt loss of fetuses after midgestation. This suggests that the absence and the overexpression of AMH may be affecting litter size via different mechanisms. Circulating AMH levels reflect the number of growing ovarian follicles, and decline to zero during ovarian ageing [34]. The phenotype of the young Amh−/− dams thus leads to the hypotheses that low levels of AMH within the ovary and/or the circulation are causal determinants of age-related decline in reproductive output. The first litters of the Amh−/− dams were smaller on average than their second litters, which adds to the evidence that the lower reproductive capacity of nulliparous dams is independent of AMH. Possible sites of anti-Müllerian hormone action AMH is potentially a complex regulator of the reproductive capacity of females, with multiple sites of action. AMH inhibits the primordial to primary follicle transition in the ovary [15] and reduces granulosa cell sensitivity to FSH in developing follicles [16]. The number of oocytes progressing to ovulation in mice is thought to be independent of AMH. However, the studies that underpin this view had group sizes [15, 35] that are insufficient to detect an average reduction in the ovulation rate of between one and two oocytes. It is therefore possible that the smaller litter size of Amh−/− dams is due to a reduction in the average number of concepti generated at mating. Equally, the antecedents of spontaneous abortion could involve the oocyte quality, with AMH potentially affecting the fitness of oocytes through its action on granulosa cells. In the Thy1.2-AMHTg/0 dams, endogenous ovarian AMH expression was normal in the developing follicles of Thy1.2-AMHTg/0 dams from the secondary stage onwards but was reduced in primary follicles. These data do not support the existence of an AMH-regulating negative feedback loop in late preantral and antral follicles but the effects on early-stage follicles require further investigation. Resorption of fetuses during spontaneous abortions involves changes in the maternal–fetal interface, enabling maternal immune cells to invade the conceptus [36]. AMHR2 is expressed by the oviduct (Pankhurst, unpublished observations), the uterus [6, 37], and by the placenta [7]. These AMH receptors are in an appropriate location to regulate implantation and the maturation of the maternal–fetal interface, but the physiological functions of these receptors has yet to be investigated. It is also unknown whether the placental and uterine AMH receptors are exclusively activated by locally produced AMH, AMH released from follicular fluids at ovulation, by circulating AMH or combinations of these sources. Successful pregnancy also requires changes in various neural networks and the pituitary. AMH is produced within the brain, and AMHR2 is broadly expressed by neurons [23]. The recent discovery that hypothalamic AMHR2 expression modulates gonadotropin-releasing hormone secretion [38] highlights the possibility that the physiological control of AMH levels within the brain and pituitary underpins successful pregnancy. Permissive versus instructive mechanisms When a trait is generated by a simple dose-response curve, no expression and overexpression are at opposite ends of a spectrum. The Amh−/− and Thy1.2-AMHTg/0 dams both have smaller litter sizes, indicating that AMH regulation of reproduction is not simple. This is not surprising given that there are multiple potential sites of action, multiple sources of AMH and the fact that AMH is part of the TGFbeta superfamily, which typically has context-dependent signaling [39]. It is premature to speculate on what the precise molecular etiology is in each of the mouse lines, as the possibilities are currently numerous. We selectively raise two points, which highlight the differences between the Amh−/− and Thy1.2-AMHTg/0 mouse lines. First, some of the known actions of AMH are dose-dependent (instructive) in vivo. However, the AMH-induced regression of the Müllerian duct is permissive, as it occurs equally in all normal male embryos, irrespective of whether their AMH levels are in the lower or upper portions of the physiological range of AMH values. If AMH has permissive actions in females, then these actions would be absent in Amh−/− dams, but should be unaffected in the Thy1.2-AMHTg/0dams, as permissive actions would be fully activated by the mouse's endogenous AMH. Second, AMH may only activate certain cell types at particular stages of the life cycle or when specific physiological or pathological stimuli are present. Sites that are normally quiescent for AMH during pregnancy would not be directly affected by the absence of Amh, but such sites may be inappropriately activated in Thy1.2-AMHTg/0females, leading to pathological disruption of pregnancy. Limitation statement Amh−/− and Thy1.2-AMHTg/0are nonphysiological extremes, and the resulting phenotypes may therefore not be indicative of normal physiology. For example, the effects of increased AMH protein synthesis in neurons have not been determined. The lack of strong behavioral deficits in the Thy1.2-AMHTg/0 mice argues against a severe detriment to neuronal function but does not exclude the possibility of a more moderate effect that could affect fertility. The definitive proof of the physiological relevance of AMH will need to involve experiments where AMH has been manipulated within the physiological range. Such experiments may need to be highly statistically powered. A reduction in litter size of less than one is relevant to the biological fitness of a mouse, but the robust detection of a decrease in litter size required a group size of 10, with each dam contributing three litters. A comparison between groups of 10 with a single litter was marginally powered (Figure 6). In this context, the extreme phenotypes of the Amh−/− and Thy1.2-AMHTg/0 mice may be of value to screen the numerous possible sites of AMH action, to identify a smaller subset of hypotheses to be tested by comparison of Amh+/+to Amh+/-mice, and/or by conditional disruption of signaling involving temporal and cell-type specificity. Conclusion Fecundity varies during the life cycle and is subject to environmental influences. The reproductive phenotype of Thy1.2-AMHTg/0 and Amh−/− mice suggest that AMH may have a broad role in the regulation of fecundity, with the levels of AMH putatively influencing whether a dam supports a conceptus to term. Supplementary data Supplementary data are available at BIOLRE online. Supplemental Table 1. Antibodies. Supplemental Figure S1. Behavioural phenotyping of the Thy1.2-AMHTg/0 mice. Supplemental Figure S2. Immunohistochemistry for AMH in the brain of Thy1.2-AMHTg/0 mice. Supplemental Figure S3. Immunohistochemistry for AMH in the spinal cord and pituitary gland of Thy1.2-AMHTg/0 mice. Acknowledgments Mrs Brandi-Lee Leathart is thanked for providing expert technical assistance. Author Contributions KK conceived and designed the transgenic mouse line; KK generated the transgene, the data generated by KK was independently verified by ISM; ISM, MP, and NB designed and performed the breeding experiments; ISM and MP analyzed the data; and ISM drafted the paper. All authors except KK approved the final version of the manuscript. † Grant Support: This study was supported by a grant from the Health Research Council (New Zealand) to ISM and MWP, grant number 14_441. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Conference Presentation: Presented in part at the 31st Annual Meeting of the European Society of Human Reproduction and Embryology, 14th–17th June 2015, Lisbon, Portugal. Edited by Dr. T. Rajendra Kumar, PhD, University of Colorado Anschutz Medical Campus. 1 The first litter of a dam tends to be smaller, with litters 2 and 3 being of similar size [30, 31]. The number of fetuses at E10 also tends to be larger than the number of pups born, as dams resorb a portion of their fetuses after E10 [30, 32]. Both of these phenomena are present in the main Amh colony: consequently, there was an a priori expectation that the three matings would be subtly different. This can be most simply controlled for by including mating number in the ANOVA. When this is done, the P value for Amh genotype was P < 0.0005. The dams did not invariably give birth after each of the first two matings (Figure 5), and the dams in the second and third matings were therefore heterogenous with respect to parity. Consequently, the differences between the matings were controlled for by using two binary variables (Nulliparous: yes/no and Stage: E10/birth). The inclusion of the additional binary variables decreases the inherent statistical power, but more accurately reflects the underlying biology. However, the observed power (0.92 at α = 0.05) and P value (0.001) for the effect of Amh genotype indicates that the sample size was sufficiently large to use this approach. References 1. MacLaughlin DT, Donahoe PK. Sex determination and differentiation. New Eng J Med  2004; 350: 367– 378. Google Scholar CrossRef Search ADS PubMed  2. McLennan IS, Pankhurst MW. Anti-Mullerian hormone is a gonadal cytokine with two circulating forms and cryptic actions. J Endocrinol  2015; 226: R45– R57. Google Scholar CrossRef Search ADS PubMed  3. Lebeurrier N, Launay S, Macrez R, Maubert E, Legros H, Leclerc A, Jamin SP, Picard JY, Marret S, Laudenbach V, Berger P, Sonderegger P et al.   Anti-Mullerian-hormone-dependent regulation of the brain serine-protease inhibitor neuroserpin. J Cell Sci  2008; 121: 3357– 3365. Google Scholar CrossRef Search ADS PubMed  4. Wang PY, Koishi K, McGeachie AB, Kimber M, Maclaughlin DT, Donahoe PK, McLennan IS. Mullerian inhibiting substance acts as a motor neuron survival factor in vitro. Proc Natl Acad Sci USA  2005; 102: 16421– 16425. Google Scholar CrossRef Search ADS PubMed  5. Clarkson AN, Talbot CL, Wang PY, Maclaughlin DT, Donahoe PK, McLennan IS. Mullerian inhibiting substance is anterogradely transported and does not attenuate avulsion-induced death of hypoglossal motor neurons. Exp Neurol  2011; 231: 304– 308. Google Scholar CrossRef Search ADS PubMed  6. Wang J, Dicken C, Lustbader JW, Tortoriello DV. Evidence for a Mullerian-inhibiting substance autocrine/paracrine system in adult human endometrium. Fertil Steril  2009; 91: 1195– 1203. Google Scholar CrossRef Search ADS PubMed  7. Novembri R, Funghi L, Voltolini C, Belmonte G, Vannuccini S, Torricelli M, Petraglia F. Placenta expresses anti-Mullerian hormone and its receptor: Sex-related difference in fetal membranes. Placenta  2015; 36: 731– 737. Google Scholar CrossRef Search ADS PubMed  8. Josso N. Paediatric applications of anti-mullerian hormone research. 1992 Andrea Prader Lecture. Hormone Res  1995; 43: 243– 248. Google Scholar CrossRef Search ADS PubMed  9. Rey RA, Lhomme C, Marcillac I, Lahlou N, Duvillard P, Josso N, Bidart JM. Anti-Müllerian hormone as a serum marker of granulosa cell tumors of the ovary: comparative study with serum alpha-inhibin and estradiol. Am J Obstet Gynecol  1996; 174: 958– 965. Google Scholar CrossRef Search ADS PubMed  10. Kissell KA, Danaher MR, Schisterman EF, Wactawski-Wende J, Ahrens KA, Schliep K, Perkins NJ, Sjaarda L, Weck J, Mumford SL. Biological variability in serum anti-Mullerian hormone throughout the menstrual cycle in ovulatory and sporadic anovulatory cycles in eumenorrheic women. Hum Reprod  2014; 29: 1764– 1772. Google Scholar CrossRef Search ADS PubMed  11. Nelson SM, Stewart F, Fleming R, Freeman DJ. Longitudinal assessment of anti-mullerian hormone during pregnancy—relationship wwith maternal adiposity, insulin, and adiponectin. Fertil Steril  2010; 93: 1356– 1358. Google Scholar CrossRef Search ADS PubMed  12. Koninger A, Kauth A, Schmidt B, Schmidt M, Yerlikaya G, Kasimir-Bauer S, Kimmig R, Birdir C. Anti-Mullerian-hormone levels during pregnancy and postpartum. Reprod Biol Endocrinol  2013; 11: 60. Google Scholar CrossRef Search ADS PubMed  13. Pankhurst MW, Clark CA, Zarek J, Laskin CA, McLennan IS. Changes in circulating proAMH and total AMH during healthy pregnancy and post-partum: a longitudinal study. PLoS One  2016; 11: e0162509. Google Scholar CrossRef Search ADS PubMed  14. Visser JA, Schipper I, Laven JS, Themmen AP. Anti-Mullerian hormone: an ovarian reserve marker in primary ovarian insufficiency. Nat Rev Endocrinol  2012; 8: 331– 341. Google Scholar CrossRef Search ADS PubMed  15. Durlinger AL, Kramer P, Karels B, de Jong FH, Uilenbroek JT, Grootegoed JA, Themmen AP. Control of primordial follicle recruitment by anti-Mullerian hormone in the mouse ovary. Endocrinology  1999; 140: 5789– 5796. Google Scholar CrossRef Search ADS PubMed  16. Visser JA, Themmen AP. Role of anti-Mullerian hormone and bone morphogenetic proteins in the regulation of FSH sensitivity. Mol Cell Endocrinol  2014; 382: 460– 465. Google Scholar CrossRef Search ADS PubMed  17. Segev DL, Hoshiya Y, Stephen AE, Hoshiya M, Tran TT, MacLaughlin DT, Donahoe PK, Maheswaran S. Mullerian inhibiting substance regulates NFkappaB signaling and growth of mammary epithelial cells in vivo. J Biol Chem  2001; 276: 26799– 26806. Google Scholar CrossRef Search ADS PubMed  18. Namkung J, Song JY, Jo HH, Kim MR, Lew YO, Donahoe PK, MacLaughlin DT, Kim JH. Mullerian inhibiting substance induces apoptosis of human endometrial stromal cells in endometriosis. J Clin Endo Metab  2012; 97: 3224– 3230. Google Scholar CrossRef Search ADS   19. Kawagishi Y, Pankhurst MW, Nakatani Y, McLennan IS. Anti-Mullerian hormone signaling is influenced by Follistatin 288, but not 14 other transforming growth factor beta superfamily regulators. Mol Reprod Dev  2017; 84: 626– 637. Google Scholar CrossRef Search ADS PubMed  20. Kano M, Sosulski AE, Zhang L, Saatcioglu HD, Wang D, Nagykery N, Sabatini ME, Gao G, Donahoe PK, Pepin D. AMH/MIS as a contraceptive that protects the ovarian reserve during chemotherapy. Proc Natl Acad Sci USA  2017; 114: E1688– E1697. Google Scholar CrossRef Search ADS PubMed  21. Durlinger AL, Gruijters MJ, Kramer P, Karels B, Ingraham HA, Nachtigal MW, Uilenbroek JT, Grootegoed JA, Themmen AP. Anti-Mullerian hormone inhibits initiation of primordial follicle growth in the mouse ovary. Endocrinology  2002; 143: 1076– 1084. Google Scholar CrossRef Search ADS PubMed  22. Caroni P. Overexpression of growth-associated proteins in the neurons of adult transgenic mice. J Neurosci Meth  1997; 71: 3– 9. Google Scholar CrossRef Search ADS   23. Wang PY, Protheroe A, Clarkson AN, Imhoff F, Koishi K, McLennan IS. Müllerian inhibiting substance contributes to sex-linked biases in the brain and behavior. Proc Natl Acad Sci USA  2009; 106: 7203– 7208. Google Scholar CrossRef Search ADS PubMed  24. Wittmann W, McLennan IS. Anti-Mullerian hormone may regulate the number of calbindin-positive neurons in the sexually dimorphic nucleus of the preoptic area of male mice. Biol Sex Differ  2013; 4: 18. Google Scholar CrossRef Search ADS PubMed  25. Wittmann W, McLennan IS. The bed nucleus of the stria terminalis has developmental and adult forms in mice, with the male bias in the developmental form being dependent on testicular AMH. Horm Behav  2013; 64: 605– 610. Google Scholar CrossRef Search ADS PubMed  26. Behringer RR, Finegold MJ, Cate RL. Mullerian-inhibiting substance function during mammalian sexual development. Cell  1994; 79: 415– 425. Google Scholar CrossRef Search ADS PubMed  27. McLennan IS, Taylor-Jeffs J. The use of sodium lamps to brightly illuminate mouse houses during their dark phases. Lab Anim  2004; 38: 1– 9. Google Scholar CrossRef Search ADS   28. Pankhurst MW, Leathart BL, Batchelor NJ, McLennan IS. The anti-Müllerian hormone precursor (proAMH) is not converted to the receptor-competent form (AMHN,C) in the circulating blood of mice. Endocrinology  2016; 157: 1622– 1629. Google Scholar CrossRef Search ADS PubMed  29. Imhoff FM, Yang D, Mathew SF, Clarkson AN, Kawagishi Y, Tate WP, Koishi K, McLennan IS. The type 2 anti-Mullerian hormone receptor has splice variants that are dominant-negative inhibitors. FEBS Lett  2013; 587: 1749– 1753. Google Scholar CrossRef Search ADS PubMed  30. Finn CA. Reproductive capacity and litter size in mice: effect of age and environment. J Reprod Fertil  1963; 6: 205– 214. Google Scholar CrossRef Search ADS PubMed  31. Johnson MS, Thomson SC, Speakman JR. Limits to sustained energy intake. III. Effects of concurrent pregnancy and lactation in Mus musculus. J Exp Biol  2001; 204: 1947– 1956. Google Scholar PubMed  32. Holinka CF, Tseng YC, Finch CE. Reproductive aging in C57BL/6J mice: plasma progesterone, viable embryos and resorption frequency throughout pregnancy. Biol Reprod  1979; 20: 1201– 1211. Google Scholar CrossRef Search ADS PubMed  33. Johnston SL, Souter DM, Erwin SS, Tolkamp BJ, Yearsley JM, Gordon IJ, Illius AW, Kyriazakis I, Speakman JR. Associations between basal metabolic rate and reproductive performance in C57BL/6J mice. J Exp Biol  2007; 210: 65– 74. Google Scholar CrossRef Search ADS PubMed  34. de Kat AC, van der Schouw YT, Eijkemans MJ, Herber-Gast GC, Visser JA, Verschuren WM, Broekmans FJ. Back to the basics of ovarian aging: a population-based study on longitudinal anti-Mullerian hormone decline. BMC Med  2016; 14: 151. Google Scholar CrossRef Search ADS PubMed  35. Visser JA, Durlinger AL, Peters IJ, van den Heuvel ER, Rose UM, Kramer P, de Jong FH, Themmen AP. Increased oocyte degeneration and follicular atresia during the estrous cycle in anti-Mullerian hormone null mice. Endocrinology  2007; 148: 2301– 2308. Google Scholar CrossRef Search ADS PubMed  36. Bonney EA, Brown SA. To drive or be driven: the path of a mouse model of recurrent pregnancy loss. Reproduction  2014; 147: R153– R167. Google Scholar CrossRef Search ADS PubMed  37. Renaud EJ, MacLaughlin DT, Oliva E, Rueda BR, Donahoe PK. Endometrial cancer is a receptor-mediated target for mullerian inhibiting substance. Proc Natl Acad Sci USA  2005; 102: 111– 116. Google Scholar CrossRef Search ADS PubMed  38. Cimino I, Casoni F, Liu X, Messina A, Parkash J, Jamin SP, Catteau-Jonard S, Collier F, Baroncini M, Dewailly D, Pigny P, Prescott M et al.   Novel role for anti-Mullerian hormone in the regulation of GnRH neuron excitability and hormone secretion. Nat Commun  2016; 7: 10055. Google Scholar CrossRef Search ADS PubMed  39. Massague J. TGFbeta signalling in context. Nat Rev Mol Cell Biol  2012; 13: 616– 630. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2017. Published by Oxford University Press on behalf of Society for the Study of Reproduction. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Biology of Reproduction Oxford University Press

Mice with either diminished or elevated levels of anti-Müllerian hormone have decreased litter sizes

Loading next page...
 
/lp/ou_press/mice-with-either-diminished-or-elevated-levels-of-anti-m-llerian-FB82VFlBIl
Publisher
Oxford University Press
Copyright
© The Author(s) 2017. Published by Oxford University Press on behalf of Society for the Study of Reproduction. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
ISSN
0006-3363
eISSN
1529-7268
D.O.I.
10.1093/biolre/iox151
Publisher site
See Article on Publisher Site

Abstract

Abstract Anti-Müllerian hormone (AMH) is both a gonadal hormone and a putative paracrine regulator of neurons, the uterus, and the placenta. A mouse line with neuronal expression of AMH (Thy1.2-AMH) was generated to examine the role of paracrine AMH in the brain. The mice had normal behavior, but unexpectantly AMH was present in the circulation of the transgenic mice. Thy1.2-AMHTg/0 studs sired pups with a normal frequency, when mated with wild-type dams. In stark contrast, Thy1.2-AMHTg/0 dams rarely gave birth, with evidence of spontaneous midgestational abortion. This leads to the hypothesis that AMH influences the capacity of dams to carry concepti to term. This hypothesis was tested by mating AMH-deficient (Amh−/−), Thy1.2-AMHTg/0, and wild-type dams when 49-, 80-, and 111 days old, using proven wild-type studs. The litter sizes from the first two matings and the number of fetuses present on the 10th day of gestation of the third mating were recorded. Thy1.2-AMHTg/0 dams carried near normal numbers of midterm fetuses, but typically produced no pups, indicating that extensive late resorption of fetuses was occurring. Amh−/− dams exhibited a lesser reduction in litter size than the Thy1.2-AMHTg/0 dams, with no evidence of enhanced loss of fetuses. In conclusion, this study provides the first evidence that high AMH levels can cause a miscarriage phenotype and that the absence of AMH affects reproductive output. Introduction Anti-Müllerian hormone (AMH) is a sexually dimorphic gonadal hormone [1, 2], which is also synthesized by various nongonadal cells, including some mature neurons [3–5], the uterus [6], and the placenta [7]. Circulating AMH is entirely derived from the gonads [8, 9], suggesting that the nongonadal sources of AMH act exclusively as paracrine regulators. In mature females, both the endocrine and paracrine sources of AMH are linked to reproduction. Circulating ovarian AMH exhibits minimal variation during the ovarian cycle [10], but declines acutely during the second and third trimester of pregnancy [11–13], and chronically with age, in parallel with the decline in the pool of growing ovarian follicles [14]. AMH slows reproductive senescence in mice [15], and modulates the influence of follicle-stimulating hormone (FSH) on granulosa cells [16]. However, the physiological actions of AMH may extend to other reproductive organs, as the AMH-specific receptor (AMHR2) is expressed in the mammary gland [17], uterus [6, 18], and placenta [7]. Speculatively, the age- and/or pregnancy-related decline in circulating AMH may modulate the transfer of resources from the mother to her offspring. Equally, the uterus and placenta synthesize AMH [6, 7], raising the possibility that the uterine and placental receptors are predominantly or exclusively activated by locally produced (paracrine) AMH. Transgenic mice with neuron-specific expression of human AMH (hAMH) were generated to study the role of paracrine AMH in the mature brain. The receptor-binding region of hAMH is 93% homologous to the murine protein and activates murine AMH receptors in multiple experimental systems [19–21]. The Thy1.2 promoter was used to delay the expression of the transgene until neurons had matured [22], thus mimicking the normal developmental onset of neuronal AMH expression, and avoiding the virilizing effect of AMH on the developing nervous system [23–25]. We report here that the Thy1.2-AMHTg/0mice contain transgenic hAMH in their circulation and exhibit a high rate of midterm loss of embryos. Materials and methods Thy1.2-AMH mice A cDNA clone containing the full-length Homo sapiens AMH mRNA was purchased from InVitrogen/Life Technologies (MGC:54308; IMAGE:5168299; Unigene Cluster ID: Hs.112432; Carlsbad, CA, USA). The AMH gene was cloned into the mouse thymus antigen 1.2 (Thy1.2) expression cassette [22] and the Thy1.2-AMH construct verified by sequencing (The Allan Wilson Centre, Massey University, Palmerston North, New Zealand). Transgenic mouse lines were then produced by microinjection of zygotes from FVB mice with the Thy1.2-AMH cDNA by AgResearch Ltd, Ruakura, Hamilton, New Zealand. The Thy1.2-AMH mouse lines were maintained by cross-breeding with Amh+/+ mice, which were the wild-type offspring from Amh+/-x Amh+/− matings, in order to generate congenic Amh and Thy1.2-AMH lines. The resulting pups were therefore either hemizygous for the transgene (Thy1.2-AMHTg/0) or wild type (Thy1.2-AMH0/0). The mice were genotyped by PCR using a forward primer located in the Thy1.2 promoter (5’-TGT AGC TTT CCC CAC CAC AGA-3΄) and a reverse primer in the AMH transgene (5’-AGG GCC TCA GTC CCC AGC AGA-3΄), with the amplified sequence being 145 bp. The Amh−/− mouse line, which carries a targeted gene disruption that globally abolishes AMH expression [26], was obtained from The Jackson Laboratory (B6;129S7-Amhtm1Bhr/J, Stock number 002187, Bar Harbor, ME, USA). All mice were housed in M.I.C.E. cages (Animal Care Systems, CO 80112, USA), as previously described [27]. The temperature was 21 ± 1°C and the relative humidity varied between 40 and 50%. The housing met all requirements specified in the National Research Council's Guide for the Care and Use of Laboratory Animals (NIH, USA) and in the guidelines of the New Zealand Ministry for Primary Industries. The experiments were approved by the Animal Ethics Committee of the University of Otago. Human anti-Müllerian hormone in the blood of Thy1.2-AMH mice The mice were terminally anaesthetized with ketamine (225 mg/kg, Phoenix Pharm) and Domitor (3 mg/kg, Medetomidine, Pfizer Animal Health), after which blood was drawn by cardiac puncture and clotted at room temperature for 1 h. The serum was aliquoted and snap-frozen in liquid nitrogen for storage at –80°C. The transgenic hAMH levels were assayed using a method that does not detect endogenous murine AMH [28]. Briefly, the samples were diluted 1:24 and were assayed using the PicoAMH ELISA (Cat# AL-124-I; Ansh Labs) as per the manufacturer's procedure, with the exception that the antibody-biotin conjugate was replaced with the antibody-biotin conjugate (100 μL) from the AMH Gen II ELISA kit (Cat# A79765; BeckmanCoulter). The picoAMH ELISA capture antibody produces high sensitivity and the Gen II detection antibody circumvents the variable ability of the picoAMH detection antibody to recognize different recombinant hAMH preparations. Quantitative reverse transcriptase-PCR detection of human anti-Müllerian hormone expression Tissue specificity The adrenal glands, brain, heart, kidney, liver, lung, skeletal muscle, and the uterus from four Thy1.2-AMHTg/0 were rapidly dissected, snap-frozen in liquid nitrogen, after which the mRNA was extracted and cDNAs generated as previously described [29]. The primers were designed to amplify the hAMH transgene but not murine Amh cDNA, with cDNA from the brains of wild-type mice used to verify primer specificity (F 5΄ AGGAAGTGACCTGGGAGCAACA 3΄, R 5΄ AGCCCAGCCCTCGTCACAGTGA 3΄). Hypoxanthine-guanine phosphoribosyltransferase (Hprt) expression (F 5΄ CTGGTGAAAAGGACC-TCTCG 3΄, R 5΄ TGAAGTACTCATTATAGTCAAGGGCA 3΄) was used to normalize data for mRNA abundance. Temporal specificity A portion of the brains from 0-, 5-, 15-, 22-, 30-, 42-, and 53-day-old Thy1.2-AMHTg/0 mice were processed as described in the previous paragraph, with the exception that glyceraldehyde 3-phosphate deydrogenase (Gapdh) expression (F 5΄ TCTTCACCACCATGGAGAAG 3΄, 5΄ ACCAAAGTTGTCATGGAT-GAC 3΄) was used to normalize the data for mRNA abundance. Localization of human anti-Müllerian hormone and endogenous anti-Müllerian hormone protein The localization of the transgenic hAMH protein was examined by immunohistochemistry using perfusion-fixed tissues, as previously described [25]. The primary antibody was a goat peptide antibody made to the N-terminal (pro region) of human MIS/AMH (#AF2748, R&D Systems, Minneapolis, MN, Supplementary Table 1). Murine endogenous AMH protein was detected with the goat polyclonal MIS-C20 antibody (Santa Cruz, Dallas, Tx) using Bouins-fixed ovaries from mice in estus. Nonimmune goat IgG and tissues from wild-type mice were used as negative controls. Observations were made using a Zeiss Axioplan microscope, equipped with 0.075, 0.75, and 1.40 numerical aperture plan-apochromat lens for bright-field observations and 0.30 and 0.50 numerical aperture neofluar lens for phase-contrast imaging. Photomicrographs were produced with a Zeiss AxioCam HRc RGB camera, with no digital correction. Breeding of the Thy1.2-AMH mice The Thy1.2-AMH mice were bred as stable stud and dam pairs, which were checked daily for births (pups or postpartum bleeding). The colony records were maintained in an in-house computer programme which includes the date of birth, date of mating, date of birth of each litter, litter size, date and cause of death. The programme uses these data to track lineages and multiple parameters relating to breeding performance, including time between mating and birth, interval between births and litter size. The first analysis of the reproduction of the Thy1.2-AMH colony was from these records, and corresponds to the period when the genetic background was being backcrossed from FVB to that of the Amh colony. Mating experiment Six Thy1.2-AMHTg/0, Thy1.2-AMH0/0 (wild-type), and Amh+/+ (wild-type) dams and 12 Amh−/− dams were paired with a proven wild-type stud for mating over a 9-day period. The dams were transferred to the stud's cage, with each stud only having one partner at any one time. Eleven wild-type studs were used (AMH+/+ mice or Thy1.2-AMH0/0 mice, derived from either of the congenic colonies), and were mated at random with the dams. The dams were mated when aged 49-, 80- and 111 days old. All pups derived from the first two matings were culled at birth, and the litter sizes recorded. During the third mating, the dams were examined daily for the presence of a postcoital plug, and euthanized 10 days after the detection of the plug. The number of fetuses being carried was recorded. Statistical analysis The difference in the frequency of births within 50 days of mating (Figure 4) were examined using a Binomial Test, with an expected frequency of 1 (frequency observed in Amh+/+ dams). The influence of genotype on the frequency of dams with no pups or fetuses was examined using a Pearson's Chi squared test, with pair-wise comparison using a Fischer exact test when a statistically significant effect of genotype was observed (P < 0.05) (Figure 5). The differences in the sizes of litters generated by Thy1.2-AMHTg/0 studs and dams were examined using a one-way ANOVA. When three groups of dams were compared, a two-way ANOVA was used: pair-wise comparisons of genotype were made using one-way ANOVA when the two-way ANOVA was significant for genotype. The litter size data were also analyzed using a three-way ANOVA approach to take parity into account; the three factors were genotype, nulliparous (yes/no), and offspring stage (i.e., pup count for litter 1 and 2 or E10 embryo for litter 3). IBM SPSS Statistics v21 and Prism 7 (GraphPad Software Inc) were used for the statistical calculations. Results Only two of the 25 founder mice contained the transgene: both were female. One founder failed to give birth after 139 days of mating, despite the stud being replaced after 60 days. The other founder gave birth to a litter of seven pups 34 days after being paired with a stud, with a single pup being born 25 days later from a second pregnancy. The dam was culled at this stage, and a colony established from the Thy1.2-AMHTg/0 mice born from the first litter. The Thy1.2-AMHTg/0 mice exhibited no gross phenotype, with the reproductive tract of female Thy1.2-AMHTg/0 being of normal size and anatomy. The Thy1.2-AMHTg/0 mice were of normal appearance and weight, and exhibited no overt neurological symptoms or atypical behaviors. Their behavior was quantitatively normal in a range of tests, including open-field, rotarod, elevated plus maze, and forced swim test, as was the response of the mice to selected neuroactive drugs (Supplemental Information File 1). The expression of hAMH mRNA was first detected in the brains of 21-day-old Thy1.2-AMHTg/0 mice, with the level of expression increasing markedly between days 30 and 42 (Figure 1). Human AMH protein was expressed in most neurons of adult Thy1.2-AMHTg/0 mice, with the notable exception of the thalamus and hypothalamus (Figure 2; Supplemental Information Files 2 and 3). In the cerebral cortex, for example, the neurons in the multiform (VI) and internal pyramidal cell (V) layers were strongly stained by the anti-AMH antibody, as were most but not all of the external pyramidal cells. There was, however, minimal or no immunoreactivity in the internal (II) and external granular (IV) layers. The transgene was also detected in a minority of pituitary cells (Supplemental File 3). The mean levels of hAMH mRNA in brain were 8.5% of housekeeping gene (Hprt) mRNA levels but were less than 0.5% in adrenal glands, heart, kidney, liver, skeletal muscle, and uterus. Slightly higher levels were observed in the lung (0.9% of Hprt mRNA levels), which is consistent with prior reports of Thy1.2 promoter-driven transgene expression [22]. Unexpectedly, serum from adult (44–98 day old) Thy1.2-AMHTg/0 mice contained 1.12 ± 0.14 nmol/L (n = 14) of hAMH. The serum of the wild-type littermates contained undetectable levels of hAMH, as expected (<0.02 pM, n = 16). AMH in the blood is a mixture of uncleaved proAMH and AMHR2-receptor-competent AMHN,C: the molecular forms of the hAMH in the blood of theThy1.2-AMHTg/0mice mirrors that of AMH in human serum [28]. The level of endogenous AMH expression in the ovary was largely similar in Thy1.2-AMHTg/0 mice when compared to Thy1.2-AMH0/0 females by immunohistochemistry (Figure 3). At most stages of follicle development, similar levels of AMH expression were observed with the exception of primary follicles which show weak staining in wild-type females but no staining in Thy1.2-AMH0/0 mice. Figure 1. View largeDownload slide The expression of the hAMH transgene with postnatal age. The level of hAMH mRNA in the brains of Thy1.2-AMHTg/0 mice were measured by quantitative PCR, and normalized to the level of Gapdh mRNA. The values for individual mice are illustrated, using blue dots for males (M) and pink squares for females (F). The number of mice at each age group was day 0, 2F, 2M; day 5, 1F, 4M; day 15, 1F, 1M; day 22 2F, 1M; day 30 3F, 2M; day 42 2M; day 53 2F. Figure 1. View largeDownload slide The expression of the hAMH transgene with postnatal age. The level of hAMH mRNA in the brains of Thy1.2-AMHTg/0 mice were measured by quantitative PCR, and normalized to the level of Gapdh mRNA. The values for individual mice are illustrated, using blue dots for males (M) and pink squares for females (F). The number of mice at each age group was day 0, 2F, 2M; day 5, 1F, 4M; day 15, 1F, 1M; day 22 2F, 1M; day 30 3F, 2M; day 42 2M; day 53 2F. Figure 2. View large Download slide hAMHTg expression in the cerebral cortex. The hAMH was detected by immunohistochemistry utilizing an antibody to human proAMH. Strong staining was observed in the cortex of Thy1.2-AMHTg/0 mice (A) but little or no staining was observed in the cortex of wild-type (Thy1.2-AMH0/0) mice (B). The locations of the cortical layers are indicated by roman numerals; layer I: molecular layer; layer II: external granular layer; layer III: external pyramidal layer; layer IV: internal granular layer; layer V: internal pyramidal layer; layer VI: multiform layer and WM; white matter. Staining can be observed in the granular neuron cell bodies of layers I and IV, the pyramidal neurons in layer V and cell bodies at all depths of layer V. The density of neurons in the molecular layer is low but the occasional AMH-positive cell can be observed in layer I, which is consistent with neuronal staining. Little or no staining was observed in the pyramidal neurons of layer III. Higher magnification images of layer II (C), layer IV-V (D), and layer VI (E) reveals staining consistent with neuronal cell bodies including the initial segment of neuronal processes in the more intensely stained cells (arrowheads). No nonspecific staining of the brain was observed with goat antibody isotype controls (Supplemental file 2). The scale bars represent 100 μm. Figure 2. View large Download slide hAMHTg expression in the cerebral cortex. The hAMH was detected by immunohistochemistry utilizing an antibody to human proAMH. Strong staining was observed in the cortex of Thy1.2-AMHTg/0 mice (A) but little or no staining was observed in the cortex of wild-type (Thy1.2-AMH0/0) mice (B). The locations of the cortical layers are indicated by roman numerals; layer I: molecular layer; layer II: external granular layer; layer III: external pyramidal layer; layer IV: internal granular layer; layer V: internal pyramidal layer; layer VI: multiform layer and WM; white matter. Staining can be observed in the granular neuron cell bodies of layers I and IV, the pyramidal neurons in layer V and cell bodies at all depths of layer V. The density of neurons in the molecular layer is low but the occasional AMH-positive cell can be observed in layer I, which is consistent with neuronal staining. Little or no staining was observed in the pyramidal neurons of layer III. Higher magnification images of layer II (C), layer IV-V (D), and layer VI (E) reveals staining consistent with neuronal cell bodies including the initial segment of neuronal processes in the more intensely stained cells (arrowheads). No nonspecific staining of the brain was observed with goat antibody isotype controls (Supplemental file 2). The scale bars represent 100 μm. Figure 3. View largeDownload slide AMH immunohistochemical staining in Thy1.2-AMH0/0 (A, B, D, F) and Thy1.2-AMHTg/0 (C, E, D) mouse ovaries. The level of AMH immunostaining is similar when follicles are compared at the preantral stage (A, C), the small antral stage (D, E), and the medium-sized antral stage (F, G). Weak AMH immunoreactivity was present in the primary follicles of Thy1.2-AMH0/0 mice (B, arrowhead) but no immunostaining was observed at this stage in Thy1.2-AMHTg/0 mice (C, arrowhead). AMH was visualized with a diaminobenzidine chromogen (brown) with a hemotoxylin counterstain (blue). Images are representative of ovaries from three mice from each genotype, collected while in estrus. The scale bar represents 100 μm. Figure 3. View largeDownload slide AMH immunohistochemical staining in Thy1.2-AMH0/0 (A, B, D, F) and Thy1.2-AMHTg/0 (C, E, D) mouse ovaries. The level of AMH immunostaining is similar when follicles are compared at the preantral stage (A, C), the small antral stage (D, E), and the medium-sized antral stage (F, G). Weak AMH immunoreactivity was present in the primary follicles of Thy1.2-AMH0/0 mice (B, arrowhead) but no immunostaining was observed at this stage in Thy1.2-AMHTg/0 mice (C, arrowhead). AMH was visualized with a diaminobenzidine chromogen (brown) with a hemotoxylin counterstain (blue). Images are representative of ovaries from three mice from each genotype, collected while in estrus. The scale bar represents 100 μm. The Thy1.2-AMHTg/0 studs and dams were mated with Amh+/+ mice to change the genetic background of the Thy1.2-AMH colony. Breeding colony data were analyzed retrospectively and the breeding performance of the Thy1.2-AMHTg/0studs and dams was shown to be distinct. When Thy1.2-AMHTg/0 studs were mated with Amh+/+ dams, all dams gave birth within 50 days of entering the stud's cage, whereas only 5 of 18 Thy1.2-AMHTg/0 dams mated with Amh+/+ studs gave birth within this time frame (Figure 4A). The litter size was also markedly smaller when the transgene was carried by the dam rather than the stud (Figure 4B), independent of the number of back crosses. The frequency of Thy1.2-AMHTg/0 pups in litters that were genotyped was close to the Mendelian ratio (49.1%, n = 242), indicating that Thy1.2-AMHTg/0 fetuses were not selectively lost in utero. The time between the stud-dam pairing and birth was independent of which sex carried the transgene (Figure 4C). Two pregnant Thy1.2-AMHTg/0dams were euthanized when they failed to give birth. They were carrying one or three healthy fetuses that were atypically large. Postmortem examination of some of the barren dams revealed the presence of fetuses that appeared to have died around midgestation. Figure 4. View largeDownload slide Breeding statistics from the Thy1.2-AMH colony, with reference to whether the dam or the stud carried the hAMH transgene. (A) The percentage of pairings involving Thy1.2-AMHTg/0studs and Thy1.2-AMHTg/0 dams that yielded a litter within 50 days of pairing with a wild-type mouse. (B) The average litter size of Thy1.2-AMHTg/0 dams and wild-type dams mated with Thy1.2-AMHTg/0 studs. (C) The average time between mating and birth. Day 0 for the first litter was the first day of pairing, whereas day zero for the second litter was the day of birth of the first litter 1. The data are the mean plus the standard error of the mean, with the number of mice indicated on the bar. The 2nd+ pairing column contains data from the second litter onwards hence the number of data points exceeds the number of stud-dam pairings. ‘#’The frequency of births by the Thy1.2-AMHTg/0is significantly less than the observed frequency (1.0) of Amh+/+ dams, mated to Thy1.2-AMHTg/0 studs: # P < 0.0005 (Binomial Test). The litter size was significantly different depending on whether the transgene was carried by the stud or the dam: *First litter P = 0.002; **Second litter P < 0.005 (one-way ANOVA). There was no significant difference in the time to give birth (first litter P = 0.76, second litter P = 0.20). Figure 4. View largeDownload slide Breeding statistics from the Thy1.2-AMH colony, with reference to whether the dam or the stud carried the hAMH transgene. (A) The percentage of pairings involving Thy1.2-AMHTg/0studs and Thy1.2-AMHTg/0 dams that yielded a litter within 50 days of pairing with a wild-type mouse. (B) The average litter size of Thy1.2-AMHTg/0 dams and wild-type dams mated with Thy1.2-AMHTg/0 studs. (C) The average time between mating and birth. Day 0 for the first litter was the first day of pairing, whereas day zero for the second litter was the day of birth of the first litter 1. The data are the mean plus the standard error of the mean, with the number of mice indicated on the bar. The 2nd+ pairing column contains data from the second litter onwards hence the number of data points exceeds the number of stud-dam pairings. ‘#’The frequency of births by the Thy1.2-AMHTg/0is significantly less than the observed frequency (1.0) of Amh+/+ dams, mated to Thy1.2-AMHTg/0 studs: # P < 0.0005 (Binomial Test). The litter size was significantly different depending on whether the transgene was carried by the stud or the dam: *First litter P = 0.002; **Second litter P < 0.005 (one-way ANOVA). There was no significant difference in the time to give birth (first litter P = 0.76, second litter P = 0.20). On the basis of the breeding performance of the Thy1.2-AMHTg/0mice, AMH was postulated to regulate litter size, at least in part by influencing whether or not a conceptus is carried to term. This hypothesis was tested with a prospective experiment that compared the reproductive output of dams with AMH over expression (Thy1.2-AMHTg/0) with wild-type dams (Thy1.2-AMH0/0, Amh+/+) and dams with a null mutation in the Amh gene (Amh−/−). The dams were mated with a proven wild-type stud for 9 days, when aged 49, 80, and 111 days. The Thy1.2-AMH0/0 and Amh+/+dams were combined as a single wild-type group, as the colonies were congenic and as none of their reproductive traits were statistically different. The output from the Thy1.2-AMHTg/0 dams in this experiment (Figure 5) mirrored the results obtained from the colony record (Figure 4). Only one litter of two pups was obtained from two matings of the six Thy1.2-AMHTg/0 dams: the number of Thy1.2-AMHTg/0dams that did not give birth (five out of six) was statistically significant from both the wild-type and Amh−/−groups (P = 0.001), which all gave birth to at least one litter and were not significantly different to each other. On the third mating, all six female Thy1.2-AMHTg/0 dams exhibited copulatory plugs, indicating that mating had occurred and four out of six were carrying fetuses on day 10 of the third mating (Figure 5). Despite the severely reduced capacity of Thy1.2-AMHTg/0 dams to produce live offspring, there was no significant effect of genotype across the three groups in the frequency of pregnancy on embryonic day 10. Figure 5. View largeDownload slide Comparison of litter sizes from three groups of dams. Each dam was mated with a proven wild-type stud for 9 days on three occasions when aged 49-, 80- and 111 days old, with the data showing the number of pups born in each of the first two litters and the number of fetuses present on the 10h day of gestation of the third pregnancy. The blue dots represent individual wild-type dams (six Amh+/+ and six Thy1.2-AMH0/0): the two wild-type groups were congenic, and were not statistically different from each other. Zero data points represent dams that either did not give birth after one of the first two pairings with a stud or were not pregnant after the third mating. *There was a statistically significant effect of genotype on the frequency of dams with no pups from the first two matings (P < 0.0005 Pearson's Chi square test). In posthoc pairwise comparisons, the Thy1.2-AMHTg/0 dams were significantly different to the wild-type group (P = 0.003) and the Amh−/− group (P < 0.0005) (two-tailed Fischer exact test). The wild-type and Amh−/− groups were not significantly different (P = 0.27). There was no significant effect of genotype on dams with the presence of at least one fetus at E10 for the third mating (P = 0.56). Figure 5. View largeDownload slide Comparison of litter sizes from three groups of dams. Each dam was mated with a proven wild-type stud for 9 days on three occasions when aged 49-, 80- and 111 days old, with the data showing the number of pups born in each of the first two litters and the number of fetuses present on the 10h day of gestation of the third pregnancy. The blue dots represent individual wild-type dams (six Amh+/+ and six Thy1.2-AMH0/0): the two wild-type groups were congenic, and were not statistically different from each other. Zero data points represent dams that either did not give birth after one of the first two pairings with a stud or were not pregnant after the third mating. *There was a statistically significant effect of genotype on the frequency of dams with no pups from the first two matings (P < 0.0005 Pearson's Chi square test). In posthoc pairwise comparisons, the Thy1.2-AMHTg/0 dams were significantly different to the wild-type group (P = 0.003) and the Amh−/− group (P < 0.0005) (two-tailed Fischer exact test). The wild-type and Amh−/− groups were not significantly different (P = 0.27). There was no significant effect of genotype on dams with the presence of at least one fetus at E10 for the third mating (P = 0.56). When the matings that resulted in pups or fetuses were analyzed, the Amh−/− dams on average carried fewer offspring (pups or fetuses) than the wild-type dams, by between 1.1 and 1.4 offspring per mating. The effect of Amh genotype was strongly significant when the three matings were examined as a group (P = 0.001), after correction for the influence of parity and whether the dam was examined at E10 or at birth (Figure 6)1. The influence of Amh genotype was also strongly evident if the analysis was restricted to the first two matings, which were examined at birth (P = 0.002, Fig 6). The low frequency of detection of offspring precluded any statistical examination of the litter size of the Thy1.2-AMHTg/0dams. Figure 6. View largeDownload slide Influence of a null mutation of Amh on reproductive output by dams. The experiment is as described in Figure 5. Dams were only included in the analysis if birth occurred (pairing 1 and 2) or if fetuses were detected at E10 from the third mating. The data is the mean plus the standard error of the mean, with the number of mice indicated on the bar. *There was a significant effect of genotype (P = 0.001) in a three-way ANOVA, controlling for whether the dam was nulliparous and whether fetuses or pups were examined.1 # There was a significant effect of genotype (P = 0.002) and parity (P = 0.008), with no significant genotype-by-parity interaction (P = 0.65) (two-way ANOVA). When the wild-type and Amh−/− dams were compared separately for each mating, the P values were as follows: first P = 0.030; second P = 0.035, third P = 0.058 (one-way ANOVA). Figure 6. View largeDownload slide Influence of a null mutation of Amh on reproductive output by dams. The experiment is as described in Figure 5. Dams were only included in the analysis if birth occurred (pairing 1 and 2) or if fetuses were detected at E10 from the third mating. The data is the mean plus the standard error of the mean, with the number of mice indicated on the bar. *There was a significant effect of genotype (P = 0.001) in a three-way ANOVA, controlling for whether the dam was nulliparous and whether fetuses or pups were examined.1 # There was a significant effect of genotype (P = 0.002) and parity (P = 0.008), with no significant genotype-by-parity interaction (P = 0.65) (two-way ANOVA). When the wild-type and Amh−/− dams were compared separately for each mating, the P values were as follows: first P = 0.030; second P = 0.035, third P = 0.058 (one-way ANOVA). Discussion The regulation of litter size is complex, both with respect to cause and mechanism. Nulliparous and older dams tend to have smaller litters, but with variation between dams, even in inbred colonies [30, 31]. Environmental influences that alter energy balance also have large effects on litter size, which can involve trade-offs between the number and size of pups [31, 33]. Mechanistically, litter size is determined by multiple factors, including the number of oocytes released, the rate of implantation, and by resorption of concepti during either the embryonic or fetal stages [30, 32]. AMH influences the rate of depletion of ovarian follicles [15], which putatively influences when a female experiences age-related difficulty to conceive. The reproductive output of the Thy1.2-AMHTg/0 and Amh−/− dams raises the possibility that AMH may also be part of the molecular mechanism(s) that enable dams to adapt their reproductive output to their current circumstances. Reproductive phenotype of Thy1.2-AMHTg/0mice The Thy1.2-AMHTg/0dams, which overexpress AMH, infrequently gave birth and when birth occurred, the litter size was atypically small, in large part due to the resorption of fetuses. The loss of fetuses was not due to an inherent lack of viability of Thy1.2-AMHTg/0 fetuses, for two reasons. First, wild-type dams successfully carried Thy1.2-AMHTg/0 fetuses sired by Thy1.2-AMHTg/0 studs to term and second, pregnant Thy1.2-AMHTg/0 dams rarely carried their wild-type fetuses to term. Ovarian overexpression of AMH causes infertility, with restriction of primordial follicle activation being the putative cause [20]. The current data do not exclude the possibility that reduced follicle activation contributes to the phenotype of Thy1.2-AMHTg/0 dams, but pregnant Thy1.2-AMHTg/0dams carried too many fetuses at midgestation for it to be a major mechanism. The influence of the transgenic AMH in the current study was of similar magnitude in the nulliparous and multiparous dams, suggesting that AMH is not an essential part of the mechanism that diminishes the size of litters of nulliparous dams. Litter size of Amh−mh dams Mice that lacked the Amh gene also had smaller litters. The magnitude of the effect was, however, smaller than that observed in the Thy1.2-AMHTg/0dams, and there was no overt loss of fetuses after midgestation. This suggests that the absence and the overexpression of AMH may be affecting litter size via different mechanisms. Circulating AMH levels reflect the number of growing ovarian follicles, and decline to zero during ovarian ageing [34]. The phenotype of the young Amh−/− dams thus leads to the hypotheses that low levels of AMH within the ovary and/or the circulation are causal determinants of age-related decline in reproductive output. The first litters of the Amh−/− dams were smaller on average than their second litters, which adds to the evidence that the lower reproductive capacity of nulliparous dams is independent of AMH. Possible sites of anti-Müllerian hormone action AMH is potentially a complex regulator of the reproductive capacity of females, with multiple sites of action. AMH inhibits the primordial to primary follicle transition in the ovary [15] and reduces granulosa cell sensitivity to FSH in developing follicles [16]. The number of oocytes progressing to ovulation in mice is thought to be independent of AMH. However, the studies that underpin this view had group sizes [15, 35] that are insufficient to detect an average reduction in the ovulation rate of between one and two oocytes. It is therefore possible that the smaller litter size of Amh−/− dams is due to a reduction in the average number of concepti generated at mating. Equally, the antecedents of spontaneous abortion could involve the oocyte quality, with AMH potentially affecting the fitness of oocytes through its action on granulosa cells. In the Thy1.2-AMHTg/0 dams, endogenous ovarian AMH expression was normal in the developing follicles of Thy1.2-AMHTg/0 dams from the secondary stage onwards but was reduced in primary follicles. These data do not support the existence of an AMH-regulating negative feedback loop in late preantral and antral follicles but the effects on early-stage follicles require further investigation. Resorption of fetuses during spontaneous abortions involves changes in the maternal–fetal interface, enabling maternal immune cells to invade the conceptus [36]. AMHR2 is expressed by the oviduct (Pankhurst, unpublished observations), the uterus [6, 37], and by the placenta [7]. These AMH receptors are in an appropriate location to regulate implantation and the maturation of the maternal–fetal interface, but the physiological functions of these receptors has yet to be investigated. It is also unknown whether the placental and uterine AMH receptors are exclusively activated by locally produced AMH, AMH released from follicular fluids at ovulation, by circulating AMH or combinations of these sources. Successful pregnancy also requires changes in various neural networks and the pituitary. AMH is produced within the brain, and AMHR2 is broadly expressed by neurons [23]. The recent discovery that hypothalamic AMHR2 expression modulates gonadotropin-releasing hormone secretion [38] highlights the possibility that the physiological control of AMH levels within the brain and pituitary underpins successful pregnancy. Permissive versus instructive mechanisms When a trait is generated by a simple dose-response curve, no expression and overexpression are at opposite ends of a spectrum. The Amh−/− and Thy1.2-AMHTg/0 dams both have smaller litter sizes, indicating that AMH regulation of reproduction is not simple. This is not surprising given that there are multiple potential sites of action, multiple sources of AMH and the fact that AMH is part of the TGFbeta superfamily, which typically has context-dependent signaling [39]. It is premature to speculate on what the precise molecular etiology is in each of the mouse lines, as the possibilities are currently numerous. We selectively raise two points, which highlight the differences between the Amh−/− and Thy1.2-AMHTg/0 mouse lines. First, some of the known actions of AMH are dose-dependent (instructive) in vivo. However, the AMH-induced regression of the Müllerian duct is permissive, as it occurs equally in all normal male embryos, irrespective of whether their AMH levels are in the lower or upper portions of the physiological range of AMH values. If AMH has permissive actions in females, then these actions would be absent in Amh−/− dams, but should be unaffected in the Thy1.2-AMHTg/0dams, as permissive actions would be fully activated by the mouse's endogenous AMH. Second, AMH may only activate certain cell types at particular stages of the life cycle or when specific physiological or pathological stimuli are present. Sites that are normally quiescent for AMH during pregnancy would not be directly affected by the absence of Amh, but such sites may be inappropriately activated in Thy1.2-AMHTg/0females, leading to pathological disruption of pregnancy. Limitation statement Amh−/− and Thy1.2-AMHTg/0are nonphysiological extremes, and the resulting phenotypes may therefore not be indicative of normal physiology. For example, the effects of increased AMH protein synthesis in neurons have not been determined. The lack of strong behavioral deficits in the Thy1.2-AMHTg/0 mice argues against a severe detriment to neuronal function but does not exclude the possibility of a more moderate effect that could affect fertility. The definitive proof of the physiological relevance of AMH will need to involve experiments where AMH has been manipulated within the physiological range. Such experiments may need to be highly statistically powered. A reduction in litter size of less than one is relevant to the biological fitness of a mouse, but the robust detection of a decrease in litter size required a group size of 10, with each dam contributing three litters. A comparison between groups of 10 with a single litter was marginally powered (Figure 6). In this context, the extreme phenotypes of the Amh−/− and Thy1.2-AMHTg/0 mice may be of value to screen the numerous possible sites of AMH action, to identify a smaller subset of hypotheses to be tested by comparison of Amh+/+to Amh+/-mice, and/or by conditional disruption of signaling involving temporal and cell-type specificity. Conclusion Fecundity varies during the life cycle and is subject to environmental influences. The reproductive phenotype of Thy1.2-AMHTg/0 and Amh−/− mice suggest that AMH may have a broad role in the regulation of fecundity, with the levels of AMH putatively influencing whether a dam supports a conceptus to term. Supplementary data Supplementary data are available at BIOLRE online. Supplemental Table 1. Antibodies. Supplemental Figure S1. Behavioural phenotyping of the Thy1.2-AMHTg/0 mice. Supplemental Figure S2. Immunohistochemistry for AMH in the brain of Thy1.2-AMHTg/0 mice. Supplemental Figure S3. Immunohistochemistry for AMH in the spinal cord and pituitary gland of Thy1.2-AMHTg/0 mice. Acknowledgments Mrs Brandi-Lee Leathart is thanked for providing expert technical assistance. Author Contributions KK conceived and designed the transgenic mouse line; KK generated the transgene, the data generated by KK was independently verified by ISM; ISM, MP, and NB designed and performed the breeding experiments; ISM and MP analyzed the data; and ISM drafted the paper. All authors except KK approved the final version of the manuscript. † Grant Support: This study was supported by a grant from the Health Research Council (New Zealand) to ISM and MWP, grant number 14_441. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Conference Presentation: Presented in part at the 31st Annual Meeting of the European Society of Human Reproduction and Embryology, 14th–17th June 2015, Lisbon, Portugal. Edited by Dr. T. Rajendra Kumar, PhD, University of Colorado Anschutz Medical Campus. 1 The first litter of a dam tends to be smaller, with litters 2 and 3 being of similar size [30, 31]. The number of fetuses at E10 also tends to be larger than the number of pups born, as dams resorb a portion of their fetuses after E10 [30, 32]. Both of these phenomena are present in the main Amh colony: consequently, there was an a priori expectation that the three matings would be subtly different. This can be most simply controlled for by including mating number in the ANOVA. When this is done, the P value for Amh genotype was P < 0.0005. The dams did not invariably give birth after each of the first two matings (Figure 5), and the dams in the second and third matings were therefore heterogenous with respect to parity. Consequently, the differences between the matings were controlled for by using two binary variables (Nulliparous: yes/no and Stage: E10/birth). The inclusion of the additional binary variables decreases the inherent statistical power, but more accurately reflects the underlying biology. However, the observed power (0.92 at α = 0.05) and P value (0.001) for the effect of Amh genotype indicates that the sample size was sufficiently large to use this approach. References 1. MacLaughlin DT, Donahoe PK. Sex determination and differentiation. New Eng J Med  2004; 350: 367– 378. Google Scholar CrossRef Search ADS PubMed  2. McLennan IS, Pankhurst MW. Anti-Mullerian hormone is a gonadal cytokine with two circulating forms and cryptic actions. J Endocrinol  2015; 226: R45– R57. Google Scholar CrossRef Search ADS PubMed  3. Lebeurrier N, Launay S, Macrez R, Maubert E, Legros H, Leclerc A, Jamin SP, Picard JY, Marret S, Laudenbach V, Berger P, Sonderegger P et al.   Anti-Mullerian-hormone-dependent regulation of the brain serine-protease inhibitor neuroserpin. J Cell Sci  2008; 121: 3357– 3365. Google Scholar CrossRef Search ADS PubMed  4. Wang PY, Koishi K, McGeachie AB, Kimber M, Maclaughlin DT, Donahoe PK, McLennan IS. Mullerian inhibiting substance acts as a motor neuron survival factor in vitro. Proc Natl Acad Sci USA  2005; 102: 16421– 16425. Google Scholar CrossRef Search ADS PubMed  5. Clarkson AN, Talbot CL, Wang PY, Maclaughlin DT, Donahoe PK, McLennan IS. Mullerian inhibiting substance is anterogradely transported and does not attenuate avulsion-induced death of hypoglossal motor neurons. Exp Neurol  2011; 231: 304– 308. Google Scholar CrossRef Search ADS PubMed  6. Wang J, Dicken C, Lustbader JW, Tortoriello DV. Evidence for a Mullerian-inhibiting substance autocrine/paracrine system in adult human endometrium. Fertil Steril  2009; 91: 1195– 1203. Google Scholar CrossRef Search ADS PubMed  7. Novembri R, Funghi L, Voltolini C, Belmonte G, Vannuccini S, Torricelli M, Petraglia F. Placenta expresses anti-Mullerian hormone and its receptor: Sex-related difference in fetal membranes. Placenta  2015; 36: 731– 737. Google Scholar CrossRef Search ADS PubMed  8. Josso N. Paediatric applications of anti-mullerian hormone research. 1992 Andrea Prader Lecture. Hormone Res  1995; 43: 243– 248. Google Scholar CrossRef Search ADS PubMed  9. Rey RA, Lhomme C, Marcillac I, Lahlou N, Duvillard P, Josso N, Bidart JM. Anti-Müllerian hormone as a serum marker of granulosa cell tumors of the ovary: comparative study with serum alpha-inhibin and estradiol. Am J Obstet Gynecol  1996; 174: 958– 965. Google Scholar CrossRef Search ADS PubMed  10. Kissell KA, Danaher MR, Schisterman EF, Wactawski-Wende J, Ahrens KA, Schliep K, Perkins NJ, Sjaarda L, Weck J, Mumford SL. Biological variability in serum anti-Mullerian hormone throughout the menstrual cycle in ovulatory and sporadic anovulatory cycles in eumenorrheic women. Hum Reprod  2014; 29: 1764– 1772. Google Scholar CrossRef Search ADS PubMed  11. Nelson SM, Stewart F, Fleming R, Freeman DJ. Longitudinal assessment of anti-mullerian hormone during pregnancy—relationship wwith maternal adiposity, insulin, and adiponectin. Fertil Steril  2010; 93: 1356– 1358. Google Scholar CrossRef Search ADS PubMed  12. Koninger A, Kauth A, Schmidt B, Schmidt M, Yerlikaya G, Kasimir-Bauer S, Kimmig R, Birdir C. Anti-Mullerian-hormone levels during pregnancy and postpartum. Reprod Biol Endocrinol  2013; 11: 60. Google Scholar CrossRef Search ADS PubMed  13. Pankhurst MW, Clark CA, Zarek J, Laskin CA, McLennan IS. Changes in circulating proAMH and total AMH during healthy pregnancy and post-partum: a longitudinal study. PLoS One  2016; 11: e0162509. Google Scholar CrossRef Search ADS PubMed  14. Visser JA, Schipper I, Laven JS, Themmen AP. Anti-Mullerian hormone: an ovarian reserve marker in primary ovarian insufficiency. Nat Rev Endocrinol  2012; 8: 331– 341. Google Scholar CrossRef Search ADS PubMed  15. Durlinger AL, Kramer P, Karels B, de Jong FH, Uilenbroek JT, Grootegoed JA, Themmen AP. Control of primordial follicle recruitment by anti-Mullerian hormone in the mouse ovary. Endocrinology  1999; 140: 5789– 5796. Google Scholar CrossRef Search ADS PubMed  16. Visser JA, Themmen AP. Role of anti-Mullerian hormone and bone morphogenetic proteins in the regulation of FSH sensitivity. Mol Cell Endocrinol  2014; 382: 460– 465. Google Scholar CrossRef Search ADS PubMed  17. Segev DL, Hoshiya Y, Stephen AE, Hoshiya M, Tran TT, MacLaughlin DT, Donahoe PK, Maheswaran S. Mullerian inhibiting substance regulates NFkappaB signaling and growth of mammary epithelial cells in vivo. J Biol Chem  2001; 276: 26799– 26806. Google Scholar CrossRef Search ADS PubMed  18. Namkung J, Song JY, Jo HH, Kim MR, Lew YO, Donahoe PK, MacLaughlin DT, Kim JH. Mullerian inhibiting substance induces apoptosis of human endometrial stromal cells in endometriosis. J Clin Endo Metab  2012; 97: 3224– 3230. Google Scholar CrossRef Search ADS   19. Kawagishi Y, Pankhurst MW, Nakatani Y, McLennan IS. Anti-Mullerian hormone signaling is influenced by Follistatin 288, but not 14 other transforming growth factor beta superfamily regulators. Mol Reprod Dev  2017; 84: 626– 637. Google Scholar CrossRef Search ADS PubMed  20. Kano M, Sosulski AE, Zhang L, Saatcioglu HD, Wang D, Nagykery N, Sabatini ME, Gao G, Donahoe PK, Pepin D. AMH/MIS as a contraceptive that protects the ovarian reserve during chemotherapy. Proc Natl Acad Sci USA  2017; 114: E1688– E1697. Google Scholar CrossRef Search ADS PubMed  21. Durlinger AL, Gruijters MJ, Kramer P, Karels B, Ingraham HA, Nachtigal MW, Uilenbroek JT, Grootegoed JA, Themmen AP. Anti-Mullerian hormone inhibits initiation of primordial follicle growth in the mouse ovary. Endocrinology  2002; 143: 1076– 1084. Google Scholar CrossRef Search ADS PubMed  22. Caroni P. Overexpression of growth-associated proteins in the neurons of adult transgenic mice. J Neurosci Meth  1997; 71: 3– 9. Google Scholar CrossRef Search ADS   23. Wang PY, Protheroe A, Clarkson AN, Imhoff F, Koishi K, McLennan IS. Müllerian inhibiting substance contributes to sex-linked biases in the brain and behavior. Proc Natl Acad Sci USA  2009; 106: 7203– 7208. Google Scholar CrossRef Search ADS PubMed  24. Wittmann W, McLennan IS. Anti-Mullerian hormone may regulate the number of calbindin-positive neurons in the sexually dimorphic nucleus of the preoptic area of male mice. Biol Sex Differ  2013; 4: 18. Google Scholar CrossRef Search ADS PubMed  25. Wittmann W, McLennan IS. The bed nucleus of the stria terminalis has developmental and adult forms in mice, with the male bias in the developmental form being dependent on testicular AMH. Horm Behav  2013; 64: 605– 610. Google Scholar CrossRef Search ADS PubMed  26. Behringer RR, Finegold MJ, Cate RL. Mullerian-inhibiting substance function during mammalian sexual development. Cell  1994; 79: 415– 425. Google Scholar CrossRef Search ADS PubMed  27. McLennan IS, Taylor-Jeffs J. The use of sodium lamps to brightly illuminate mouse houses during their dark phases. Lab Anim  2004; 38: 1– 9. Google Scholar CrossRef Search ADS   28. Pankhurst MW, Leathart BL, Batchelor NJ, McLennan IS. The anti-Müllerian hormone precursor (proAMH) is not converted to the receptor-competent form (AMHN,C) in the circulating blood of mice. Endocrinology  2016; 157: 1622– 1629. Google Scholar CrossRef Search ADS PubMed  29. Imhoff FM, Yang D, Mathew SF, Clarkson AN, Kawagishi Y, Tate WP, Koishi K, McLennan IS. The type 2 anti-Mullerian hormone receptor has splice variants that are dominant-negative inhibitors. FEBS Lett  2013; 587: 1749– 1753. Google Scholar CrossRef Search ADS PubMed  30. Finn CA. Reproductive capacity and litter size in mice: effect of age and environment. J Reprod Fertil  1963; 6: 205– 214. Google Scholar CrossRef Search ADS PubMed  31. Johnson MS, Thomson SC, Speakman JR. Limits to sustained energy intake. III. Effects of concurrent pregnancy and lactation in Mus musculus. J Exp Biol  2001; 204: 1947– 1956. Google Scholar PubMed  32. Holinka CF, Tseng YC, Finch CE. Reproductive aging in C57BL/6J mice: plasma progesterone, viable embryos and resorption frequency throughout pregnancy. Biol Reprod  1979; 20: 1201– 1211. Google Scholar CrossRef Search ADS PubMed  33. Johnston SL, Souter DM, Erwin SS, Tolkamp BJ, Yearsley JM, Gordon IJ, Illius AW, Kyriazakis I, Speakman JR. Associations between basal metabolic rate and reproductive performance in C57BL/6J mice. J Exp Biol  2007; 210: 65– 74. Google Scholar CrossRef Search ADS PubMed  34. de Kat AC, van der Schouw YT, Eijkemans MJ, Herber-Gast GC, Visser JA, Verschuren WM, Broekmans FJ. Back to the basics of ovarian aging: a population-based study on longitudinal anti-Mullerian hormone decline. BMC Med  2016; 14: 151. Google Scholar CrossRef Search ADS PubMed  35. Visser JA, Durlinger AL, Peters IJ, van den Heuvel ER, Rose UM, Kramer P, de Jong FH, Themmen AP. Increased oocyte degeneration and follicular atresia during the estrous cycle in anti-Mullerian hormone null mice. Endocrinology  2007; 148: 2301– 2308. Google Scholar CrossRef Search ADS PubMed  36. Bonney EA, Brown SA. To drive or be driven: the path of a mouse model of recurrent pregnancy loss. Reproduction  2014; 147: R153– R167. Google Scholar CrossRef Search ADS PubMed  37. Renaud EJ, MacLaughlin DT, Oliva E, Rueda BR, Donahoe PK. Endometrial cancer is a receptor-mediated target for mullerian inhibiting substance. Proc Natl Acad Sci USA  2005; 102: 111– 116. Google Scholar CrossRef Search ADS PubMed  38. Cimino I, Casoni F, Liu X, Messina A, Parkash J, Jamin SP, Catteau-Jonard S, Collier F, Baroncini M, Dewailly D, Pigny P, Prescott M et al.   Novel role for anti-Mullerian hormone in the regulation of GnRH neuron excitability and hormone secretion. Nat Commun  2016; 7: 10055. Google Scholar CrossRef Search ADS PubMed  39. Massague J. TGFbeta signalling in context. Nat Rev Mol Cell Biol  2012; 13: 616– 630. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2017. Published by Oxford University Press on behalf of Society for the Study of Reproduction. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com

Journal

Biology of ReproductionOxford University Press

Published: Jan 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off