Defective mammopoiesis, but normal hematopoiesis, in mice with a targeted disruption of the prolactin gene
1997; Springer Nature; Volume: 16; Issue: 23 Linguagem: Inglês
10.1093/emboj/16.23.6926
ISSN1460-2075
Autores Tópico(s)Cytokine Signaling Pathways and Interactions
ResumoArticle1 December 1997free access Defective mammopoiesis, but normal hematopoiesis, in mice with a targeted disruption of the prolactin gene Nelson D. Horseman Corresponding Author Nelson D. Horseman Department of Molecular and Cellular Physiology, University of Cincinnati, Cincinnati, OH, 45267-0576 USA Division of Endocrinology and Metabolism, University of Cincinnati, Cincinnati, OH, 45267-0576 USA Search for more papers by this author Wenzhu Zhao Wenzhu Zhao Department of Molecular and Cellular Physiology, University of Cincinnati, Cincinnati, OH, 45267-0576 USA Search for more papers by this author Encarnacion Montecino-Rodriguez Encarnacion Montecino-Rodriguez Department of Pathology and Laboratory Medicine and the Jonsson Comprehensive Cancer Center, UCLA School of Medicine, Los Angeles, CA, 90095-1732 USA Search for more papers by this author Minoru Tanaka Minoru Tanaka Department of Molecular and Cellular Physiology, University of Cincinnati, Cincinnati, OH, 45267-0576 USA Department of Biochemistry, Mie University School of Medicine, Mie, Japan Search for more papers by this author Kunio Nakashima Kunio Nakashima Department of Biochemistry, Mie University School of Medicine, Mie, Japan Search for more papers by this author Sandra J. Engle Sandra J. Engle Department of Molecular and Cellular Physiology, University of Cincinnati, Cincinnati, OH, 45267-0576 USA Search for more papers by this author Frost Smith Frost Smith Division of Comparative Pathology, University of Cincinnati, Cincinnati, OH, 45267-0576 USA Search for more papers by this author Edith Markoff Edith Markoff Division of Pediatric Endocrinology, University of Cincinnati, Cincinnati, OH, 45267-0576 USA Search for more papers by this author Kenneth Dorshkind Kenneth Dorshkind Department of Pathology and Laboratory Medicine and the Jonsson Comprehensive Cancer Center, UCLA School of Medicine, Los Angeles, CA, 90095-1732 USA Search for more papers by this author Nelson D. Horseman Corresponding Author Nelson D. Horseman Department of Molecular and Cellular Physiology, University of Cincinnati, Cincinnati, OH, 45267-0576 USA Division of Endocrinology and Metabolism, University of Cincinnati, Cincinnati, OH, 45267-0576 USA Search for more papers by this author Wenzhu Zhao Wenzhu Zhao Department of Molecular and Cellular Physiology, University of Cincinnati, Cincinnati, OH, 45267-0576 USA Search for more papers by this author Encarnacion Montecino-Rodriguez Encarnacion Montecino-Rodriguez Department of Pathology and Laboratory Medicine and the Jonsson Comprehensive Cancer Center, UCLA School of Medicine, Los Angeles, CA, 90095-1732 USA Search for more papers by this author Minoru Tanaka Minoru Tanaka Department of Molecular and Cellular Physiology, University of Cincinnati, Cincinnati, OH, 45267-0576 USA Department of Biochemistry, Mie University School of Medicine, Mie, Japan Search for more papers by this author Kunio Nakashima Kunio Nakashima Department of Biochemistry, Mie University School of Medicine, Mie, Japan Search for more papers by this author Sandra J. Engle Sandra J. Engle Department of Molecular and Cellular Physiology, University of Cincinnati, Cincinnati, OH, 45267-0576 USA Search for more papers by this author Frost Smith Frost Smith Division of Comparative Pathology, University of Cincinnati, Cincinnati, OH, 45267-0576 USA Search for more papers by this author Edith Markoff Edith Markoff Division of Pediatric Endocrinology, University of Cincinnati, Cincinnati, OH, 45267-0576 USA Search for more papers by this author Kenneth Dorshkind Kenneth Dorshkind Department of Pathology and Laboratory Medicine and the Jonsson Comprehensive Cancer Center, UCLA School of Medicine, Los Angeles, CA, 90095-1732 USA Search for more papers by this author Author Information Nelson D. Horseman 1,2, Wenzhu Zhao1, Encarnacion Montecino-Rodriguez3, Minoru Tanaka1,4, Kunio Nakashima4, Sandra J. Engle1, Frost Smith5, Edith Markoff6 and Kenneth Dorshkind3 1Department of Molecular and Cellular Physiology, University of Cincinnati, Cincinnati, OH, 45267-0576 USA 2Division of Endocrinology and Metabolism, University of Cincinnati, Cincinnati, OH, 45267-0576 USA 3Department of Pathology and Laboratory Medicine and the Jonsson Comprehensive Cancer Center, UCLA School of Medicine, Los Angeles, CA, 90095-1732 USA 4Department of Biochemistry, Mie University School of Medicine, Mie, Japan 5Division of Comparative Pathology, University of Cincinnati, Cincinnati, OH, 45267-0576 USA 6Division of Pediatric Endocrinology, University of Cincinnati, Cincinnati, OH, 45267-0576 USA *Corresponding author. E-mail: [email protected] The EMBO Journal (1997)16:6926-6935https://doi.org/10.1093/emboj/16.23.6926 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Prolactin (PRL) has been implicated in numerous physiological and developmental processes. The mouse PRL gene was disrupted by homologous recombination. The mutation caused infertility in female mice, but did not prevent female mice from manifesting spontaneous maternal behaviors. PRL-deficient males were fertile and produced offspring with normal Mendelian gender and genotype ratios when they were mated with heterozygous females. Mammary glands of mutant female mice developed a normal ductal tree, but the ducts failed to develop lobular decorations, which is a characteristic of the normal virgin adult mammary gland. The potential effect of PRL gene disruption on antigen-independent primary hematopoiesis was assessed. The results of this analysis indicated that myelopoiesis and primary lymphopoiesis were unaltered in the mutant mice. Consistent with these observations in PRL mutant mice, PRL failed to correct the bone marrow B cell deficiency of Snell dwarf mice. These results argue that PRL does not play any indispensable role in primary lymphocyte development and homeostasis, or in myeloid differentiation. The PRL−/− mouse model provides a new research tool with which to resolve a variety of questions regarding the involvement of both endocrine and paracrine sources of PRL in reproduction, lactogenesis, tumorigenesis and immunoregulation. Introduction Prolactin (PRL) is closely associated with the stimulation of lactogenesis in mammals, but its secretion in non-mammalian species, and the multiple effects caused by administration of exogenous PRL (Nicoll and Bern, 1972) have led to suggestions that PRL may play important roles in systems other than the mammary gland. Recent discoveries have reinforced the notion that PRL may be fundamental to the development of the reproductive (Cooke, 1995) and hematopoietic (Hooghe-Peters and Hooghe, 1995; Koojiman et al., 1996) systems. The fact that the PRL receptor is expressed in cells in the testis, prostate gland, seminal vesicles and ovary (Nicoll and Bern, 1972; Cooke, 1995) is consistent with a role for the hormone in reproduction. Similarly, the PRL receptor has been detected on up to 95% of bone marrow cells and thymocytes (Dardenne et al., 1991; O'Neal et al., 1991; Gagnerault et al., 1993; Touraine et al., 1994), suggesting a potential role in hematopoiesis. The PRL receptor is a member of the type I cytokine receptor superfamily, which includes numerous hematopoietic growth factor receptors (Kelly et al., 1993; Horseman and Yu-Lee, 1994). However, despite reports that PRL can potentiate effector function of lymphoid and myeloid cells during secondary immune responses (Gala, 1991; Matera et al., 1991; Kelley et al., 1992; Sabharwal et al., 1992; Murphy et al., 1993; Hooghe-Peters and Hooghe, 1995; Warwick et al., 1995; Koojiman et al., 1996), its involvement in the primary, antigen-independent development and homeostasis of these populations is not resolved. Our current understanding of the physiology and pathobiology of PRL is based on administration of exogenous hormones to animals or cell cultures, and ablation of pituitary function by surgery, drugs or spontaneous genetic mutations that suppress pituitary PRL secretion. PRL is synthesized in many tissues other than the pituitary gland (Ben-Jonathan et al., 1996). Pituitary ablation cannot remove potentially important local sources of PRL in some tissues. Consequently, many long-standing controversies regarding the role of PRL in key developmental and physiological processes have remained unresolved. To circumvent various limitations in our understanding of PRL actions, a strain of mice with a targeted disruption of the PRL gene was generated and analyzed. The results of these analyses demonstrate that PRL is required for normal reproduction and mammary gland development in adult females; but PRL is not required for somatic growth, male reproduction, spontaneous maternal behavior, or primary differentiation or homeostatic maintenance of the blood cells. Results Generation of PRL-deficient mice The PRL gene was mutated by a targeted insertion of a neomycin resistance gene (PGK–neo) into the region encoding the second α helix of the PRL protein (Figure 1A). Targeting was confirmed by genomic Southern blotting using enzyme cuts that were both 5′ (BpmI) and 3′ (HindIII) of the targeting site (Figure 1B). PCR amplification of genomic DNA yielded the predicted 420 and 930 bp fragments from the normal and targeted gene (Figure 1B). Pituitary mRNA was assayed by reverse transcriptase–PCR to confirm the loss of the PRL gene product at the mRNA level. Amplification of the 5′ portion of the PRL mRNA (encoded by exons 1–3) yielded an expected 276 bp product in +/+, +/− and −/− mice. Primers for the 3′ portion of the mRNA (encoded by exons 4 and 5) yielded the expected 312 bp product in +/+ and +/− mice, and no product in the −/− mice. These results are consistent with the prediction that the mice should express a 3′-truncated PRL transcript as a consequence of the PGK–neo insertion. Figure 1.Targeted disruption of the mouse PRL gene. (A) The organization of the wild-type mouse PRL gene is shown in relation to the targeting vector. The relative positions of the exons and introns are depicted along with restriction enzyme sites that were used for preparing the vector, probes and diagnostic fragments. Restriction enzyme sites are labeled as follows: A, AlwN1, B, BpmI, H, HindIII, S, SacI, Xc, XcmI, Xh, XhoI. The targeted locus is described in the third line with two diagnostic probes. (B) Genomic Southern blotting of wild-type (lanes 1) and targeted (lanes 2) DNA following digestion with either BpmI (left panel) or HindIII (right panel) and probing with outside probe 1 or probe 2 (see A), respectively. The sizes of the hybridizing bands are labeled at the right margin of the gels. The map of the wild-type locus predicts bands of 6.5 and 8.6 kb with BpmI and HindIII, respectively, and the map of the targeted locus predicts bands of 8.1 and 10.2 kb with BpmI and HindIII, respectively. A probe for the neomycin cassette detected only the upper (8.1 and 10.2 kb) bands in each blot (data not shown). Genotyping of five mice by PCR is depicted in the bottom panel. Primers for genotyping consisted of 5′ oligos directed to intron C and to the neo gene, and a 3′ oligo directed to exon 4. These primers produced a 420 bp product from the wild-type allele and a 930 bp product from the targeted allele under the conditions used for amplification. The size markers are labeled in the left-most lane. Lanes 1 and 2 were DNA from wild-type, lanes 3 and 4 were heterozygous, and lane 5 was DNA from a homozygous mutant mouse. Download figure Download PowerPoint Western blotting of pituitary gland extracts was used to assess the status of PRL gene expression at the protein level (Figure 2). Purified mouse PRL migrated as a 25 kDa monomer band with a larger (50 kDa) band that presumably represents a PRL dimer. A band identical to 25 kDa PRL was present in extracts from male and female +/+ mice. Heterozygous mice synthesized predominantly the 25 kDa product; and female heterozygotes, which synthesized higher PRL levels than males, also had a detectable amount of immunoreactive protein which migrated at 11 kDa. Homozygous male and female mice synthesized exclusively the 11 kDa immunoreactive PRL product. The PGK–neo cassette insertion truncated the pre-PRL polypeptide at serine 117 in the second α-helix and added a 12-amino-acid extension (sequence: PRL117–TDPPGCRNSIS–stop) before the first stop codon in PGK–neo. This results in a 129-amino-acid precursor that would be cleaved to 100 residues by removing the signal peptide. The calculated molecular weight of the polypeptide predicted from the sequence of the targeted gene was 11 306 Da, which corresponds to the 11 kDa immunoreactive band in the PRL−/− mice. Under the conditions of these Western blots, where 10 μg of pituitary extracts were loaded, small amounts of immunoreactive protein migrated faster than the main bands. These smaller products may represent proteolytically cleaved PRL. Figure 2.Western blot analysis of pituitary proteins. Pituitary extracts (10 μg) were immunoblotted as described in Materials and methods. Molecular weight markers are depicted at the left margin. The first lane is control mouse PRL (3 μg). The sex and genotypes of mice from which each extract was prepared are labeled at the top of each lane. Both wild-type and heterozygous mouse pituitaries contained similar amounts of the full-length immunoreactive PRL. Homozygous mutant mice produced only the predicted truncation product at a molecular weight of ∼11 kDa. Download figure Download PowerPoint Bioassay of pituitary PRL Pituitary extracts from homozygous wild-type and heterozygous mice had indistinguishable levels of PRL bioactivity (Table I), The average level of PRL in male mice pituitaries was ∼12% of that in female mice pituitaries. PRL bioactivity was completely undetectable in pituitaries from both male and female −/− mice. Table 1. PRL bioactivity in mouse pituitary extracts Genotype [PRL] (ng/μg protein) ± SD and (n)a Male +/+ 2.8 ± 0.4 (4) Male +/− 3.9 ± 0.7 (4) Male −/− <0.001 (4) Female +/+ 28.3 ± 1.7 (3) Female +/− 28.6 ± 5.9 (5) Female −/− <0.001 (3) An Nb2 cell bioassay was performed on lysates of pituitary glands of individual mice. Each lysate was assayed in duplicate. The minimum detectable level of PRL bioactivity in the assay was 1 pg. a n = number of individual animals in each group. Fertility, maternal behavior and somatic growth of PRL−/− mice Heterozygous crosses yielded the expected ratios of genotypes and genders in the offspring (Table II). There was no measurable difference in fetal survival of −/− mice and their littermates. Heterozygous females produced normal litter sizes and had no problems with nursing their offspring. Table 2. Mating records for heterozygous pairs +/− (M)×+/− (F) +/+ +/− −/− Male 77 (33.6%) 109 (47.6%) 43 (18.8%) Female 42 (21.4%) 112 (57.2%) 42 (21.4%) Total 119 (28%) 221 (52%) 85 (20%) Each cell of the table records the number and frequency of offspring with the genotypes identified at the top of the columns. Homozygous PRL−/− female mice were completely infertile. Female PRL−/− mice were mated with male mice of known fertility and no litters were produced following more than 15 matings. Each female mated repeatedly at irregular intervals, without entering a state of pseudopregnancy. Estrous cycles of females were assessed by vaginal smear cytology (Champlin, 1973) in six PRL−/− mice. The females all underwent cycles that displayed all of the phases, but the patterns were very irregular. Unlike normal female mice, which have only a single proestrus and a single estrus day in each cycle, the PRL−/− mice had cycles with multiple days of proestus and/or multiple days of estrus. Individual females did not establish any consistent pattern of cycling; subsequent cycles could be either longer or shorter than the normal 4–5 day cycle, with no predictable cycle length for any individual. Nulliparous female mice (8 weeks old) were tested for maternal behaviors (pup retrieval and crouching) toward foster pups. Four of six +/+ females, five of seven +/− females and six of seven −/− females retrieved 2-day-old foster pups and crouched over them in a nursing position within 30 min after placement of the pups in the home cage. Thus, PRL deficiency does not cause any profound defect in the spontaneous maternal behavior exhibited by laboratory mice. In contrast to the reproductive abnormalities of the female −/− mice, males with the disrupted PRL gene were fully fertile. Matings between −/− male and +/− female mice produced normal litter sizes and normal 1:1 Mendelian gender and genotype ratios (Table III). Table 3. Mating records for homozygous-null male×heterozygous female matings −/− (M)×+/− (F) +/− −/− Male 60 (50.4%) 59 (49.6%) Female 67 (51.9%) 62 (48.1%) Total 127 (51%) 121 (49%) Each cell of the table records the total number and frequency (in parentheses) of male or female offspring with the genotypes identified at the top of the columns. Gender and genotype distributions did not vary significantly from the Mendelian 1:1 ratios (χ2 test). Somatic growth of the mice was not significantly affected by disruption of the PRL gene. Figure 3 shows growth curves for both male and female mice through the first 6 weeks. The −/− mice grew normally, increasing to approximately triple their weight between 2 and 6 weeks of age. There was no significant difference between the body weights of control and targeted mice at any of the ages. In addition to the growth curves done on the young mice, −/− mice were examined at 6 months old, at which age the males averaged 38.9 g and the females averaged 28.6 g. Therefore, PRL deficiency had no discernible effect on growth of the mice at any age. Figure 3.Growth curves for normal and PRL-deficient mice. The mean body weight (± standard deviation) is depicted for each group. The number of mice in each group was at least 17. Download figure Download PowerPoint Gross pathology and histopathology at necropsy No macroscopic lesion was found in either male or female −/− mice, nor was there any difference at the gross level between control and PRL−/− mice. Of the wide array of tissues examined (see Materials and methods), the only histological abnormality in the 6-week-old −/− mice was a subtle effect in the pituitary glands, where there was a decrease in the volume of the acidophilic cells, consistent with deficient PRL biosynthesis and storage. As there was no bioactive PRL detected in the pituitary glands of −/− mice, pituitary gland function has not yet been studied in any more detail. Further analysis of the pituitary glands will require development of specific N-terminal and C-terminal antibodies that can differentially detect the wild-type and the targeted PRL polypeptides. Histological examination of the mammary glands at 6 weeks of age was equivocal, showing only some modest signs of hyperplasia and neutrophil infiltration in the −/− female mice. Abnormal mammary gland development in adult virgin PRL−/− mice Mammary gland development includes the formation of a branched ductal system that is decorated with terminal and lateral lobules in virgin adult mice (Figure 4A). The growth and differentiation of the mammary gland is under the control of several hormones whose precise roles are still unclear. In PRL−/− mice, terminal end buds form during puberty and the ductal tree grows normally (Figure 4B). However, in adult PRL−/− mice, the mammary gland ductal system grows into an extended branching network that is completely devoid of either terminal or lateral lobulation (Figure 4C). In normal mice, the differentiation of the ductal system results in a compact glandular structure which undergoes full lobuloalveolar development during pregnancy (Figure 4D). In PRL−/− mice, the mammary ducts ended as blunt tubes, or extended tapered tubes, without any lateral or terminal decorations (Figure 4E and F). Despite the lack of lobulation, there was no compensatory increase in the number of branches formed during the development of the mammary gland. Figure 4.Mammary gland development is defective in adult PRL-deficient mice. Whole mounts of abdominal mammary glands were stained with safronin O (A, B, E and F) or iron hematoxylin (C and D). (A) Mammary gland of 5-month-old normal (PRL+/−) mouse showing the extensive decoration of the ductal tree by lobulations (white arrows). (B) Pubescent (6-week-old) PRL−/− mouse mammary gland. The branching ducts (black arrow) and terminal end buds (white arrow) are indistinguishable in normal and PRL-deficient mouse mammary glands at this age. (C) Mammary gland from 5-month-old PRL−/− littermate of the mouse in (A). Note the nakedness of all of the ducts (black arrow) and termination of some ducts by end buds (white arrow). (D) Lobuloalveolar development of midpregnant PRL+/− mouse mammary gland Development of the glandular system is more extensive during pregnancy that in the virgin adult (A). (E) Magnification of normal (PRL+/−) virgin adult (5 month) mammary gland showing the development of lobulations associated with the ductal system (black arrows). (F) Magnification of PRL−/− virgin adult (5 month) mammary ductal system showing termination of ducts as tapered tubes (black arrow), blunt-ending tubes (arrowhead) or terminal end buds (white arrow). Download figure Download PowerPoint Effects of PRL on lymphopoiesis In view of speculation that PRL is a lymphopoietic hormone, B and T cell development was assessed in PRL−/− mice. The data in Figure 5A and Table IV indicate that the frequency of CD45R+sIgM− and surface IgM+ (sIgM+) cells is comparable in the knockout mice and their +/− littermates. In order to assess the potential of PRL to affect development in a minor subpopulation of developing B lineage cells, the expression of CD43, CD45R, HSA and IgM was used to resolve bone marrow B lineage cells into CD45R+CD43+HSA− pro-B cells (Fraction A), CD45R+CD43+HSA+ progenitors (Fraction B + C), CD45R+CD43−HSA+ pre-B cells (Fraction D) and CD45R+sIgM+ B cells (Fraction E and F). The data in Figure 6 demonstrate that the frequency of cells in these various B lineage fractions is comparable between the PRL−/− mice and their PRL+/− littermates. As bone marrow cellularity in PRL−/− mice and their PRL+/− littermates was nearly identical (Table IV), there was no significant difference in the absolute number of B lineage cells in the mice. Figure 5.Representative FACScan profiles of (A) B lineage cells in the bone marrow of PRL+/− and PRL−/− mice and (B) of CD4- and CD8–expressing cells in the thymus of PRL+/− and PRL−/− mice. Download figure Download PowerPoint Figure 6.Frequency of B lineage cells in fractions A–F in PRL−/− (open bars; n = 12) and PRL+/− (filled bars; n = 6) mice. Download figure Download PowerPoint Table 4. Bone marrow B lymphopoiesis and myelopoiesis in prolactin knockout mice Genotype Cells ×10−6 CD45R+IgM− IgM+ CD11b (Mac–1) CFU-GM/5×104 +/−a 18.7 ± 6.7c 22.4 ± 5.2 8.9 ± 5.3 51.7 ± 5.3 147.5 ± 38.0d −/−b 13.9 ± 6.8 19.1 ± 6.8 6.2 ± 3.3 57.3 ± 11.3 162.0 ± 17.0d All values are given as mean ± SD. a n = 6. b n = 12. c Cell counts based on analysis of two femurs and two tibia per mouse. d Assays run on four mice in each group. The analysis of PRL effects on B cell development in Snell dwarf (dw/dw) mice supports the above findings. As previously reported and confirmed in Table V, the frequency of CD45R+ B lineage cells in the bone marrow of these mice is significantly depressed when compared with levels present in their +/? littermates (Murphy et al., 1992; Montecino-Rodriguez et al., 1996, 1997). The data in the table indicate that while treatment of dw/dw mice with 100 μg of ovine PRL/day for 2 weeks resulted in an increase in the number of bone marrow cells, the frequency of CD45R+sIgM− and sIgM+ B lineage cells remained depressed. However, as reported previously (Montecino-Rodriguez et al., 1996), B cell defects in the bone marrow of dw/dw mice can be corrected by thyroxine. Table 5. Prolactin treatment does not stimulate bone marrow B lymphopoiesis in dwarf mice Treatment (n)a Cells ×10−6 Percentage CD45R+, sIgM− Percentage IgM+ Saline (8) dw/dw 7.1 ± 2.3 5.2 ± 4.6 7.4 ± 5.1 PRL (10) dw/dw 9.3 ± 1.9b? 5.5 ± 3.8 4.2 ± 1.8 GH (4) dw/dw 17.2 ± 8.4c 7.3 ± 1.1 4.6 ± 0.7 GH + PRL (4) dw/dw 15.6 ± 5.9c 8.5 ± 4.2 4.6 ± 0.6 Thyroxine (5) dw/dw 10.0 ± 2.8 22.9 ± 4.1d 8.9 ± 1.2 All values are given as mean ± SD. a Number in parenthesis represents the number of mice in each group. b Value significantly different from saline-treated group (P < 0.025). c Value significantly different from saline-treated group (P < 0.005). d Value significantly different from saline-treated group (P < 0.0025). The potential role of PRL in primary T cell development was determined by comparing the number and frequency of CD4- and/or CD8-expressing thymocytes in PRL−/− mice and their PRL+/− littermates. As shown in Figure 5B and Table VI, the frequencies of CD4−CD8− double negative, CD4+CD8+ double positive and single positive CD4 and CD8 thymocytes were comparable in PRL−/− mice and their PRL+/− littermates. However, although not statistically significant, the average number of cells in the thymus of the PRL+/− mice was lower than in the PRL−/− mice. Table 6. Thymus cellularity and frequency of thymocyte subpopulations in PRL knockout mice Genotype Cells ×10−6 Percentage of cells expressing: CD4−8− CD4+8+ CD4+ CD8+ +/−a 82.7 ± 16.8 2.0 ± 0.4 81.5 ± 4.0 13.1 ± 3.0 3.5 ± 1.7 −/−b 129.7 ± 41.8 1.6 ± 0.3 83.0 ± 3.1 12.5 ± 2.3 2.9 ± 0.8 All values are given as mean ± SD. a n = 5. b n = 9. Cellularity of secondary lymphoid tissues in PRL−/− mice The data in Table VII demonstrate that spleen and lymph node cellularity and the frequency of IgM-, CD4-, CD8- and CD11b (Mac-1)-expressing cells in PRL−/− mice and their PRL+/− littermates were comparable. The mean number of sIgM+ cells was lower in the spleen of PRL−/− mice than in their PRL+/− littermates, although not significantly so. Table 7. Frequency of B and T cells in the spleen (SPL) and lymph nodes (LN) Organ Genotype (n)a Cells ×10−6 Percentage of cells expressing: IgM+ CD4+ CD8+ CD11b (Mac−1) SPL +/− (6) 143.3 ± 30.7 39.7 ± 8.1 29.8 ± 1.9 14.7 ± 3.1 7.6 ± 3.1 SPL −/− (12) 93.5 ± 39.5 31.6 ± 7.7 30.7 ± 4.6 16.5 ± 3.5 6.8 ± 2.8 LN +/− (4) 1.2 ± 0.2 3.9 ± 1.2 63.5 ± 1.3 30.5 ± 1.3 9.3 ± 2.4 LN −/− (4) 1.4 ± 0.3 3.8 ± 2.1 63.2 ± 5.5 30.2 ± 4.7 8.6 ± 1.2 All values are given as mean ± SD. Lymph node cellularity is given per single lymph node and the phenotypic analysis was carried out on cells pooled from two axillary and two inguinal lymph nodes. a The number of mice in each analysis is given in parentheses. Myelopoiesis in PRL−/− mice The expression of the PRL receptor on the majority of hematopoietic cells also raised the possibility that PRL was required for normal development of myeloid cells. Therefore, the frequency of granulocyte–macrophage progenitors responsive to GM-CSF and of CD11b (Mac-1)-expressing cells were enumerated in the PRL−/− mice. The data in Table IV demonstrate that no differences in the frequency of CFU-GM and CD11b (Mac-1)+ cells between the PRL−/− mice and their littermates were observed. The hematocrits (51.1 ± 1.5 in PRL−/− and 53.0 ± 0.14 in PRL+/− mice) were similar as well. Discussion While PRL has been demonstrated to be required during lactogenesis in the pregnant mammal, its importance in reproductive function in the non-pregnant female has been poorly understood. In addition, there has been speculation that PRL might play a key role in the development of the hematopoietic system. One difficulty in addressing these issues has been the lack of an appropriate animal model for assessing PRL function. There have been three approaches to studying the effects of PRL deficiency in animals: surgical hypophysectomy (Bates et al., 1962), pharmacological inhibition of PRL secretion (Nagy et al., 1983) and analysis of pituitary dwarf models (Murphy et al., 1992). None of these approaches has been completely satisfactory, for two reasons. First pituitary ablation does not inactivate potentially important PRL synthesis in extrapituitary tissues and, secondly, many of the changes in these animals might be due to other hormone or drug effects rather than PRL deficiency, per se. To circumvent some of these limitations, we chose to produce a mouse model with an isolated PRL deficiency. The aim of the studies reported herein was to assess the effects of PRL deficiency during postnatal development, with particular emphasis on mammary gland development, reproductive capacity, somatic growth and hematopoiesis. PRL deficiency in these mice was produced by using homologous recombin
Referência(s)