| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
John Curtin School of Medical Research (T.A., D.G.S., C.G.V., C.C.G.), The Australian National University, and The Canberra Hospital and The Australian National University Medical School (T.A., D.G.S.), Canberra ACT 2601, Australia; Monash Institute of Medical Research and The Australian Research Council Centre of Excellence in Biotechnology and Development (M.K.O., C.K., D.M.d.K.), Monash University, Melbourne, Victoria 3168. Australia; Garvan Institute of Medical Research (A.S., N.J.L.), Sydney, New South Wales 2010, Australia; Murdoch Childrens Research Institute (S.S.M.M.), Royal Children’s Hospital, Melbourne, Victoria 3052, Australia; Phenomix Australia Pty Ltd (K.N., C.L.), and The Australian Phenomics Facility (C.C.G.), Canberra ACT 2601, Australia; and Flinders Medical Centre and Flinders University (N.P.), Bedford Park, Adelaide, South Australia 5042, Australia
Address all correspondence and requests for reprints to: Dr. Todor Arsov, Immunogenomics Laboratory, John Curtin School of Medical Research, Australian National University, Mills Road, P.O. Box 334, Canberra ACT 2601, Australia. E-mail: todor.arsov{at}anu.edu.au.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Two independent groups reported that mutations in human Alms1 are responsible for this syndrome in humans (11, 12, 13, 14, 15, 16). Alms1 is a widely expressed gene that encodes a 461-kDa [4169-amino acid (aa)] protein with a largely unknown function (15, 16). It contains a large tandem repeat domain coded by a single exon 8 comprising 34 imperfect repeats of a 44-aa sequence. This domain constitutes 40% of the protein. ALMS1 also has a short polyglutamine segment (17 Glu repeats, aa 13–29), which is followed by seven alanine residues (aa 30–36). In addition to a predicted leucine zipper domain (aa 2480–2501), this protein has an evolutionary conserved motif, the Alms motif, near the C terminus. This sequence is similar to a sequence of a predicted protein of unknown function in mouse and macaque (15, 16).
Although ALMS1 has been found to localize to centrosomes and basal bodies of cell cilia, its function remains largely unknown (17, 18). Based on the clinical resemblance of Alström syndrome to the Bardet-Biedl syndrome and data showing that cilia play a sensory as well as mechanical role in several cell types, it has been hypothesized that basal body and/or ciliary dysfunction may underpin the pathogenesis of Alström syndrome (18). However, the way in which such a cellular abnormality could cause obesity is unclear. Disordered appetite regulation could be central to such a disorder, but metabolic studies in children with Alström syndrome have, to date, been inconclusive. Further, because of the complex interaction between general well-being and fertility, the nature of the reproductive abnormalities in this syndrome remains uncertain.
Recently an Alms1 gene-trapped deleted mouse has been described (19) in which partial deletion of the mouse Alms1 gene caused a phenotype similar to the human condition and implicated ALMS1 in intracellular trafficking. Although diabetes is a constant and serious complication in both male and female Alström syndrome patients and in the fat aussie mice described herein, only male Alms1 gene-trapped mice developed moderate diabetes, and this appeared to be relatively late in onset. Although hypogonadism was reported in Alms1 gene-trapped mice, no data were provided on their reproductive performance, testicular weight, or reproductive hormone levels (19).
Here we describe the identification and phenotypic characterization of a new mouse model of Alström syndrome, the fat aussie (foz/foz) mouse, bearing a spontaneous 11-bp deletion (foz) in exon 8 of the mouse Alms1 gene. This mutation leads to a phenotype that recapitulates the human condition and demonstrates direct involvement of Alms1 in the regulation of spermatogenesis.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
), a recently described gene that is widely expressed and disrupted by mutations in a human obesity syndrome, Alström syndrome (15, 16). Sequencing of Alms1 cDNA from foz/foz liver revealed an 11-bp deletion in exon 8 at nucleotides 3918–3928, compared with the parental NOD sequence (Fig. 1, B and C). Similar to several human mutations, the fat aussie deletion causes a frame-shift and premature termination codon within this domain, eliminating the C-terminal two thirds of the protein (8, 9, 10, 11, 12, 13, 14, 15). PCR using primers flanking the deletion confirmed absolute concordance between obesity and homozygosity for the deletion (Fig. 1D).
|
). Similar obesity occurred in male Alms1 foz/foz animals (Fig. 3A). There was no significant difference in BMI between heterozygous mice and +/+ controls at any age (Figs. 2A and 3A).
|
|
and 3C). In addition, older mice had significantly higher values for lean body weight (Figs. 2D and 3D).
Measurements of food intake in individually housed male and female mice showed that food consumption was indistinguishable in Alms1 foz/foz mice compared with wild-type controls at age 30 d (data not shown). However, both male and female mutant mice became hyperphagic relative to wild-type mice from 60 d of age. At age 180 d, Alms1 foz/foz mice ate 30–42% more than their heterozygous and wild-type littermates (Figs. 2B and 3B). Male foz/foz mice consumed 54 kJ/d compared with 39 kilojoules (kJ)/d for their normal siblings (P < 0.001; Fig. 3B), whereas the daily energy intake of female Amls1 foz/foz animals was 53 kJ/d compared with 40 kJ/d for controls (P < 0.001; Fig. 2B).
Metabolic Changes in Fat Aussie Mice
After the development of obesity, Alms1 foz/foz mice developed the characteristic features of type 2 diabetes. Fed with a normal diet, one of eight Alms1 foz/foz males became hyperglycemic by age 75 d (defined as having a fasting blood glucose concentration greater than 10 mmol/liter and glycosuria, with an average glycemia of 10.4 mmol/liter), four of eight were diabetic by age 90 d (range of glycemia 10.1–18.4 mmol/liter), six of eight were diabetic by age 135 d (range of glycemia 14.6–36.2 mmol/liter), and all were diabetic with glucose levels above 10 mmol/liter and glycosuria by 150 d of age. None of the heterozygotes and +/+ siblings developed hyperglycemia or glycosuria (Fig. 5A). Female Alms1 foz/foz developed diabetes slightly later: one of 11 was diabetic by age 105 d (with an average glycemia of 10.4 mmol/liter), nine of 11 were diabetic by age 120 d (range of glycemia, 12.1–18.6 mmol/liter), and nine of 11 were diabetic by age 135 d (range of glycemia, 10.5–23.2 mmol/liter) (Fig. 5B). Before developing overt hyperglycemia, Alms1 mutant (foz/foz) mice were glucose intolerant as judged by the delayed peak of glycemia and longer time needed to return to baseline levels after ip administration of glucose (Fig. 5A).
|
, B–D), indicating a compensatory ß-cell growth response. Many islets formed central cystic spaces. Collectively, these data indicate a compensatory hyperinsulinemia for insulin resistance in the mutant mice, which failed to maintain normal glycemia in older mice.
|
, A and B). This was accentuated by high-fat feeding: cholesterol levels in mutant mice fed with a high-fat diet were up to 200% higher than cholesterol levels in controls (P < 0.01; Fig. 5, C and D). Serum triglyceride concentrations remained normal in mutant mice, even when fed a high-fat diet (data not shown). Both male and female mutant mice had enlarged, pale livers, and histologic examination revealed moderate to severe steatosis (data not shown).
Male Infertility in Fat Aussie Alms1 Mutant Mice Is Attributable to a Primary Defect in Spermatogenesis
All attempts to breed Alms1 foz/foz males (25 mice, age 60–180 d) were unsuccessful whereas heterozygote males produced normal sized litters (data not shown). Testes from male Alms1 foz/foz mice were nearly one third the size and weight of those from heterozygotes or wild-type mice (Fig. 6, A and B). Histological analysis identified that there was an overall decrease in germ cells at all stages of spermatogenesis beyond the spermatogonial stage. Of those germ cells that progressed through meiosis, the vast majority disappeared at step 8–9 of spermiogenesis, namely in the conversion of the round to elongating spermatids (Fig. 6, C and D). The occasional almost mature spermatid was seen in the epithelium, but these must have been reabsorbed as spermatozoa were not found in the epididymis (data not shown). The germ cell loss was consistent with the increase in TUNEL-positive cells indicative of apoptosis (Fig. 6, E and F).
|
, A and B). Significant numbers of spermatids showed two nuclei with acrosomal caps surrounded by a continuous plasma membrane (Fig. 7C), suggesting a failure of cytokinesis. The tails of the few elongated spermatids that were observed showed abnormal increases in the number of outer dense fibers (>9) in the midpiece of spermatids, and their shape and size varied considerably in contrast to spermatids from a comparable stage in wild-type mice (Fig. 7, D and E). The microtubule arrangements in the axoneme were usually normal, but occasionally, variations in the numbers or placement of doublet microtubules were found (Fig. 7, D and E). The mitochondria of spermatids showed bizarre profiles with a marked increase in the electron density and sparse inner mitochondrial membranes (Fig. 7, E and F). It was not possible to determine whether the mitochondrial phenotype was due to a developmental defect or as a consequence of degeneration.
|
). This was in striking contrast to well-defined localization of acetylated tubulin to elongated spermatid tails of wild-type testes (Fig. 6G). Cells within the testicular interstitium appeared normal, an observation that was consistent with normal serum testosterone levels. These observations, in combination with normal FSH and inhibin levels (data not shown) in foz/foz mice, strongly suggested that male infertility was not secondary to a failure of the hypothalamo-pituitary-gonadal axis, but due to a primary testicular defect. The possibility remains, however, that spermatogenesis was further compromised in older mice after the development of diabetes and a decline in general health. Young female Alms1 foz/foz mice were fertile, although litter sizes were smaller than those of corresponding heterozygotes and wild type mice. After development of obesity, female mutant mice were infertile. At this stage, their ovaries contained primary and secondary follicles but were devoid of corpora lutea, indicating an anovulatory state (data not shown).
Audiological Findings
Because deafness is often a presenting sign in human Alström syndrome, and hearing function in the inner ear depends on functional hair cell cilia, we conducted an auditory-evoked brainstem response potential analysis in fat aussie mice. At the age of 150 d Alms1 foz/foz and +/+ mice had comparable hearing thresholds [26 ± 1.5 decibels (dB) vs. 27.5 ± 4.0 dB; P > 0.05]; however, older Alms1 foz/foz mice (360 d of age) had significantly increased hearing threshold (67.1 ± 6.1 dB vs. 40.8 ± 6.0 dB; n = 6–7 mice per group; P < 0.05). All of the Alms1 foz/foz mice (n = 7) had a hearing threshold of at least 50 dB, three mice had a threshold of at least 80 dB. The findings of reduced hearing in older mutant mice are consistent with those reported in the Alms1 gene-trapped model (19).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Like human Alström syndrome, Alms1 foz/foz mice developed insulin resistance and compensatory hyperinsulinemia. With increasing age both male and female mice developed type 2 diabetes. This spontaneous mutant model of Alström syndrome is similar to the human syndrome with respect to the development of diabetes (1) and differs from Alms1 gene-trapped mice (19) in which only male mice develop diabetes. The phenotypic difference in the mouse models may relate to either mouse interstrain differences or to the hypomorphic nature of the Alms1 gene-trapped allele (19).
In the present model we observed a massive expansion of pancreatic islets and interestingly a formation of unusual cystic cavities in Alms1 foz/foz mice. The relevance of these cystic cavities is unknown but may be a specific primary feature of loss of Alms1 function in islets of Langerhans. Hypercholesterolemia, but not hypertriglyceridemia, is a conspicuous metabolic abnormality in Alms1 foz/foz mice, although the reasons for this are unclear. The possibility arises that Alms1 may play additional roles in cellular lipid partitioning (e.g. fatty acid transport and lipoprotein assembly and secretion), but this requires further investigation. In support of this hypothesis, we have recently noted that high-fat intake exacerbates hepatic steatosis in the Alms1 foz/foz mouse (23); this may provide insight into the pathogenesis of nonalcoholic fatty liver disorders, and more detailed studies of hepatic lipid partitioning in this model are underway.
The infertile state of female mice may be linked to obesity and the related changes in insulin resistance, because obese women with polycystic ovarian disease show infertility (24). Further studies are required to explore this potential link.
Our data demonstrate that the infertility in male Alms1 foz/foz mice was due to a spermatogenic disruption. This view is supported by the normal levels of FSH, immunoreactive inhibin, and testosterone. Although there was a generalized loss of germ cells, there was a more severe decrease in the conversion of round to elongated spermatids. However, it is important to note that spermatids can, on rare occasions, generate a tail structure virtually completing spermatogenesis, indicating that the distal centriole of the spermatid can generate an axoneme. This is consistent with an earlier report that Alms1 gene-trapped mice assemble cilia normally (19) and our observations of apparently normal auditory function at a young age.
The failure of the majority of germ cells to complete the conversion of a round to an elongating spermatid is of interest. It is at this stage that the manchette forms around the nucleus of the spermatid and the sperm tail/axoneme nucleates (25). The manchette is a microtubular structure that is hypothesized to be involved in nuclear elongation and in the transport of proteins involved in outer dense fiber formation in the tail (25). Similarly, the assembly of the sperm tail in general requires transport of proteins from the cytoplasm of spermatids down the cytoplasmic canal or sheath surrounding the axoneme of the sperm tail in what is presumably a microtubule-dependent process. The specific stage of this arrest coupled with the observation of numerous examples of a failure of cytokinesis at meiosis, abnormal outer dense fiber shape and number in elongated spermatids, and the presence of aberrantly expressed acetylated tubulin in Sertoli cells is strongly suggestive of abnormal microtubular dynamics. For the later observation, however, it cannot be ruled out that the presence of acetylated tubulin within Sertoli cells is a consequence of germ cell phagocytosis.
ALMS1 is a component of centrosomes and the basal bodies of cilia (17, 18, 19). A single nonmotile primary cilium is present on most cells in the body, where localization of specific receptors suggests it serves as a sensory device or antenna (26, 27, 28, 29). Primary cilia are found on fat-storing hepatocytes (30, 31, 32, 33), pancreatic ß-cells (34, 35), and are prominent on leptin-responsive hypothalamic neurons that control satiety (36, 37, 38, 39, 40, 41). The primary neuronal cilium is considered to be a relict of ancestral neurons—a nonfunctional remnant of the Cnidarian nerve net (the skin brain). The cilia from these cells have a distinct function in sensing pheromones, food, and fluid flow. Primary neuronal cilia represent an active region of the neuron containing ATP-requiring mechanisms to transport molecules, including potential signals, inside the cell (40). Some receptors, such as somatostatin 3 receptors and 5-hydroxytryptamine 6 receptors, are selectively concentrated on these surface structures (41). It is therefore tempting to hypothesize that primary neuronal cilia have a role in sensing or tuning peripheral satiety signals. Similarly, dysfunction of the primary cilia on pancreatic endocrine cells and their ductal cell precursors may cause hyperproliferation (34, 35), leading to the large, cystic islets and hyperinsulinemia in fat aussie. The tail of sperm cells is a motile flagellum related to motile primary cilia, providing a more readily understood link between the Alms1 mutation and defects in development of motile spermatozoa.
In summary, we have identified mice with a spontaneous mutation in the Alms1 gene. Alms1 foz/foz mice of both genders exhibit hyperphagia associated with profound obesity, insulin resistance, diabetes with morphologic changes in pancreatic islets, and several features of metabolic syndrome. Male fat aussie mice are sterile due to primary spermatogenic abnormalities. The present observations in a spontaneous mouse model of Alström syndrome therefore provide insight into some of the biological functions of ALMS1 and how these could be related to the pathogenesis of human obesity, diabetes, and infertility.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Mapping and Sequencing
Chromosomal linkage of the foz mutation was established in an F2 intercross between NOD and C57Bl/10 mice using a panel of 88 polymorphic short tandem repeat markers spaced at 20–30 centimorgans throughout the mouse genome. Fine mapping was done using F3 recombinants and additional polymorphic markers spaced at 1 centimorgan within the candidate interval (D6Mit17, D6Mit209, D6Mit281, D6Mit29, D6Mit213, and D6Mit102). Total RNA from foz/foz and NOD liver was prepared using TRIzol (Invitrogen Life Technologies, Carlsbad, CA) and reverse transcribed using random hexamers. Alms1 cDNA was PCR amplified and sequenced with primers designed to be complementary to the NCBI Alms1 transcript.
Screening Mice for the foz Mutation
A mutation-specific PCR using primers flanking the deletion (forward, ACA ACT TTT CAT GGC TCC AGT; reverse, TTG GCT CAG AGA CAG TTG AAA) was designed to screen mice for the presence of foz mutation (Fig. 1D). The PCR conditions were 95 C for 5 min, followed by 35 cycles of 95 C, 55 C, and 72 C for 1 min each and 72 C for 10 min. PCR products were visualized on 4% agarose gel.
Food Consumption
Food consumption studies were carried out in individually housed mice. Measurements were performed on 3 consecutive days. Each of the analyzed groups (both sexes and three genotypes) included six mice at 180 d of age.
BMI Analysis
Weight and nasal-anal length data were monitored from 30–180 d of age. Nasal-anal length was measured to the nearest millimeter in anesthetized mice. BMI values were calculated according to the formula 100*weight (in grams)/length (in cm2) and expressed in 100 g/cm2 units. Each of the analyzed groups (both sexes and three genotypes) included 10–12 mice.
Glucose Tolerance Test
ip Glucose tolerance test (2 mg glucose/g body weight) was carried out in mice fasted for 18 h. Glucose levels were determined at time points 0, 15, 30, 90, 120, and 180 min using an Advantage II glucometer (Roche, Indianapolis, IN).
DXA Scanning
DXA scans were performed in young (70–90 d old) and old (250 d old) male and female mice fed on chow (five mice in each genotype group) as described previously (42). Mice were killed immediately before placing them in the animal holder in the DXA scanner (PIXImus mouse densitometer, GE Medical Systems Lunar, Madison, WI).
Laboratory Analyses
Overnight fasted blood samples were obtained by retroorbital bleeding at different time points. Serum lipid profiles were determined using standard clinical chemistry procedures (Department of Clinical Chemistry, The Canberra Hospital). Random blood glucose levels were monitored at regular 15-d intervals using an Advantage II glucometer (Roche, Indianapolis, IN). Insulin levels were measured by commercial ELISA (Mercodia, Uppsala, Sweden), serum total testosterone concentrations were measured with a mouse-specific kit (ICN Biomedicals, Costa Mesa, CA), and serum FSH and inhibin levels were determined as described previously (43, 44).
Histological Analysis
Testes were fixed in Bouin’s fixative overnight and processed into paraffin using standard methods. Sections (5 µm) were stained with periodic acid Schiff’s reagent and counterstained with hematoxylin. Other organs were fixed in 10% neutral-buffered formalin and stained with hematoxylin and eosin using standard methods.
Testis Immunohistochemistry
Testis sections were stained for acetylated tubulin using monoclonal mouse antiacetylated tubulin Ig (clone 6–11B; Sigma Chemical Co., St. Louis, MO) at a dilution of 1:5000 and the DAKO EnVision kit as recommended by the manufacturer (DAKO Corp., Carpinteria, CA). Sections were counterstained with hematoxylin. Cells undergoing apoptosis were visualized using the Apotag kit TUNEL assay as recommended by the manufacturer (Intergen Co., Purchase, NY).
Testis Electron Microscopy
Testes from wild-type and foz/foz animals were fixed and processed for electron microscopy as described previously (45) with the exception that testes were decapsulated before fixation.
Auditory Brain Stem Response
Click-evoked auditory brainstem response thresholds were determined using a Navigator Pro diagnostic system (Bio-logic Systems Corp., Mundelain, IL) as described previously (46).
Statistical Analyses
Results are presented as mean ± SEM. Comparisons between the groups using parametric (Bonferroni’s multiple comparison test) or nonparametric (Dunn’s multiple comparison test) one-way ANOVA (PRISM version 4.0; GraphPad Software, Inc., San Diego, CA). When standard skewness and kurtosis values were within the –2 to +2 range, distributions were considered normal. P values < 0.05 were considered statistically significant.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Conflict of Interest Statement: The findings presented in the paper are original and have not been previously submitted for publication; all authors concur with the submission, and none have any conflict of interest with publication of these findings.
First Published Online March 2, 2006
1 C.C.G. and N.P. contributed equally as senior authors to this paper. 
Abbreviations: aa, Amino acids; BMI, body mass index; dB, decibel; DXA, dual x-ray absorptiometry; NOD, nonobese diabetic.
Received for publication December 7, 2005. Accepted for publication February 22, 2006.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  L. B. Scheinfeldt, S. Biswas, J. Madeoy, C. F. Connelly, E. E. Schadt, and J. M. Akey Population Genomic Analysis of ALMS1 in Humans Reveals a Surprisingly Complex Evolutionary History Mol. Biol. Evol., June 1, 2009; 26(6): 1357 - 1367. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M.-A. Cornier, D. Dabelea, T. L. Hernandez, R. C. Lindstrom, A. J. Steig, N. R. Stob, R. E. Van Pelt, H. Wang, and R. H. Eckel The Metabolic Syndrome Endocr. Rev., December 1, 2008; 29(7): 777 - 822. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
