Skin Histology. Dorsal skin sections from HF control (A) and KO (B) mice were fixed, paraffin embedded, sectioned at 10 m m and 

Context in source publication

Context 1

... is not usually observed in normal mice until after the age of 18 months [28]. The failure to thrive could not be attributed to inadequate food intake for, surprisingly, the KO mice ate significantly more food than the HF controls, when normalized for body weight ( Fig. 2 panel C). Neither could the condition be ascribed to a defect in digestion, since the protein and lipid content of the feces was not significantly different in the two groups (Table 1). Although grip strength of the KO mice was significantly lower than that of HF controls (Fig. 2, panel D), the KO mice performed as well as the HF controls on the accelerating rotarod, which assesses motor coordination and balance for a short interval, typically 1–2 minutes for each of three tests (Fig. 2, panel E). However, when the rotarod was maintained at constant 5 rpm, in a procedure designed to assess endurance, the KO mice fared less well (these experiments were performed only with female mice). HF control mice were able to complete 3 hours on the device without difficulty, averaging only one fall every 144 min, and remained lively at the end of the experiment. In contrast, KO mice quickly became exhausted and fell repeatedly, necessitating termination of the experiment after 15 min. Thus while balance and motor coordination appear unaffected by the knockout, stamina was considerably reduced. When evaluated singly in a spontaneous-activity open field test, the KO mice were less inclined to explore the environment and adopted the inquisitive rearing posture less frequently than did the HF control mice. Many of the KO mice walked with splayed rear legs and dragged their posterior on the ground. Shivering was commonly observed. Overall activity of the KO mice in the open field test was much lower than HF controls (Fig. 2, panel F). When placed in an open field as a group, HF control mice immediately separated and began investigating the area. In contrast, the KO mice showed no interest in exploring their environment and remained huddled together, as if seeking warmth and indeed body temperature measurements confirmed that these KO mice suffered from hypothermia (Fig. 2, panel G). KO mice of either sex that were not treated with tamoxifen did not develop any of the phenotypic changes described above. A survey of the literature on the use of the tamoxifen-inducible B6.Cg-Tg(cre/Esr1)5Amc/J strain in the absence of a floxed target-gene did not uncover any reports of long-term phenotypic changes resembling those identified in our study. Thus, in order to minimize the number of animals required for the study, we considered it unnecessary to include mice carrying the Cre gene but lacking the floxed Mcat gene. It has been reported that tamoxifen administration to mice carrying the inducible Cre can lead to transient cardiomyopathy and changes in energy metabolism, even in mice lacking a floxed transgene [29]. However, this condition is fully reversible within one month of tamoxifen withdrawal. Since the phenotypic changes observed in our KO mice were observed only after tamoxifen-treatment, were not evident until several months after cessation of tamoxifen treatment and were persistent for the lifespan of the animals, we can confidently attribute the changes specifically to the disruption of the Mcat gene. The most conspicuous feature of the phenotype revealed by gross examination at necropsy was the almost complete absence of white adipose tissue by 7 months post-tamoxifen treatment. Indeed mature visceral and subcutaneous white adipose could only be identified on histological examination. In contrast, there was no significant loss of brown adipose tissue (Fig. S2), which plays an important role in thermogenesis rather than energy storage and originates from a different cell lineage than does white adipose [30]. With the exception of the spleens of a sub-population of KO mice, which were considerably larger than those of control HF mice, there was no statistically significant difference in absolute weights of other organs between KO and HF mice (Fig. S2). Histological examination of skin removed from KO mice by biopsy revealed a severely depleted subcutaneous adipose layer as well as hyperkeratinization (Fig. 3). It is well recognized that cycling of the hair follicle through phases of intense mitotic activity (anagen), regression (catagen) and dormancy (telogen) is accompanied by expansion and contraction of the subcutaneous adipose layer. New evidence indicates that adipose precursor cells within the dermal fat layer are directly responsible for driving activation of follicular stem cells and induction of the anagen phase, probably by the expression and secretion of platelet-derived growth factor [31]. Other studies have uncovered a correlation between atrophy of subcutaneous adipose tissue and hair loss [32,33]. Although we cannot rule out the possibility that specific biochemical defects in hair follicles and/or sebaceous glands contribute to hair loss, it seems plausible that in our KO mouse model, the loss of subcutaneous fat plays a role in the development of alopecia. Clearly, loss of the insulation normally afforded by both hair and subcutaneous adipose must place an additional burden on the energy requirements of the KO mice and contribute to the development of hypothermia. We did not notice any unusual morphologic changes in muscle, bone, or connective tissue in the spinal column, or hind limbs that might account for the observed kyphosis, poor posture or abnormal gait of the KO animals. Nevertheless, it remains possible that subtle changes may have occurred that were not readily apparent in routine sections stained with Hematoxylin and Eosin. Twenty percent of male and 30% of female KO mice were anemic, exhibiting elevated red blood cell distribution widths, low hemoglobin levels, low red cell counts, increased mean corpuscu- lar volume and elevated reticulocyte levels (Fig. 4, A–E). Reticulocytes derived from anemic KO mice were larger than were those from non-anemic KO and HF control mice as is typically found when intense stimulation of erythropoiesis results in the premature release of ‘stress’ reticulocytes into the circulation [34]. In contrast, only minor differences were observed in white cell levels between KO and HF control mice (Table S1). Analysis of peripheral blood smears from anemic KO mice confirmed the presence of enlarged red blood cells and also revealed the presence of codocytes. These abnormal erythrocytes typically result from a decreased content of hemoglobin that manifests as a pallid ring flanking central and peripheral hemoglobinized zones (Fig. 4, panel F). The anemic KO mice corresponded to the subpopulation exhibiting enlarged spleens (Fig. S2) and showed evidence of lymphoid atrophy and prominent extramedullary hematopoiesis. Predominantly myeloid hematopoiesis was noted in the bone marrow from anemic KO mice. Siderocytes, which contain iron granules not associated with hemoglobin and can be indicative of abnormal erythropoiesis, were absent from spleen and bone marrow in these mice. Nine of the ten anemic mice also developed severe rectal prolapse, a condition not observed in either the non-anemic KO or HF control animals (Fig. 4, panel G). Histological examination of the prolapse revealed signs of neutrophil invasion and inflammation (details not shown). Replacement of bedding with filter paper revealed blood spots originating from the prolapsed anemic KO mice but not from either the non-anemic KO or HF control animals (Fig. 4, panel H). We were unable to detect blood in feces from prolapsed anemic KO mice indicating that the blood spots originated from the inverted rectum. Biotin labeling experiments revealed that the lifespan of red cells was reduced in the anemic KO mice (Fig. 4, panel I). Based on the calculated half-lives of 7.5 d and 18 d for red cells in the anemic KO and HF control mice, respectively, and assuming that blood volume amounts to , 7% of body mass, we calculated that the loss of red cells in the prolapsed, anemic mice exceeded that expected from normal turnover by , 60 m L per day. There was no evidence of erythrophagocytosis in any tissue that might indicate increased removal of defective or senescent erythrocytes; neither were plasma bilirubin levels elevated in the anemic KO mice (Fig. 4, panel J), as would be expected for a hemolytic anemia. Thus the anemia likely results primarily from chronic blood loss. Indeed mice subjected to repeated bleeding develop a macrocytic anemia similar to that described here [35]. Nevertheless, since red blood cell distribution width, a parameter that normally rises with age and is a strong predictor of mortality in humans [36], is greater in both anemic and non-anemic KO mice than in HF mice, we cannot unequivocally rule out the possibility that a defect in erythropoiesis also contributes to the anemia. Blood chemistry analysis revealed that none of the markers for liver function (alkaline phosphatase, alanine amino transferase, aspartate amino transferase and bilirubin), kidney function (blood urea nitrogen, creatinine and electrolytes) or pancreatic function (amylase) were significantly altered in ...