SEARCH

SEARCH BY CITATION

Keywords:

  • Epididymis;
  • principal cells;
  • sperm counts;
  • lanthanum nitrate;
  • claudins

Abstract

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgement
  7. References

ABSTRACT: Past studies have shown that the epithelial lining of the epididymis in adult mice deficient in protective protein cathepsin A (PPCA −/−) becomes swollen and vacuolated as a result of an accumulation of pale lysosomes, some very large, in addition to the presence of an abundance of macrophages infiltrating the intertubular spaces. The purpose of this study was to assess the integrity of the epididymal epithelial—blood barrier in these altered mice by characterizing the distribution of claudins (Cldns) and the leakiness of tight junctions to lanthanum nitrate. A second goal was to characterize sperm motility behavior in PPCA −/− mice using computer-assisted sperm analyses (CASA). The results indicated that lanthanum nitrate penetrated apical junctional complexes between adjacent epithelial cells and entered the epididymal lumen in PPCA −/− mice but not in control PPCA +/+ mice. Immunostaining for Cldns 1, 3, 8, and 10 revealed unique patterns of expression based on cell type and region specificity in PPCA +/+ mice, which were much different in PPCA –/– mice. PPCA –/– mice showed reduced intensities of immunoreactions, complete absence of immunoreactions, and appearance of atypical cytoplasmic immunoreactions. CASA indicated that sperm counts in the PPCA –/– mice were 70% reduced, and among other problems, there was a fourfold higher percentage of static sperm in PPCA –/– mice compared with controls. These results suggest that PPCA deficiency causes structural changes to the blood-epididymal barrier as evidenced by lanthanum nitrate and Cldns expression that affects the luminal environment of the epididymis, resulting in altered sperm motility.

Protective protein cathepsin A (PPCA) is a lysosomal carboxypeptidase that forms a fully functional and stable high molecular weight multi-enzyme complex with α-neuraminidase and β-galactosidase (d'Azzo et al, 1982; Galjart et al, 1991). PPCA has 2 functions. First, it facilitates the intracellular routing, lysosomal targeting, andactivationof α-neuraminidase, and second, it protects both β-galactosidase and α-neuraminidase against rapid proteolytic degradation inside lysosomes (d'Azzo et al, 1982; Galjart et al, 1991; Zhou et al, 1996; van der Spoel et al, 1998). PPCA is sorted to lysosomes as a 54-kd precursor protein via the mannose-6-phosphate receptor—mediated pathway, where it is processed into its mature 32/20-kd, disulfide-linked 2-chain form (Jackman et al, 1992; Hanna et al, 1994; Itoh et al, 1995; Rottier et al, 1998). The creation of a mouse model deficient in PPCA allowed a means to investigate the importance of this enzyme in a wide variety of cells and tissues of the body. The absence of PPCA leads to the storage of sialylated oligosaccharides and glycopeptides within lysosomes and their accumulation in a wide variety of cells and tissues of the body (van Pelt et al, 1988; d'Azzo et al, 1995; Zhou et al, 1995; Sohma et al, 1999), in addition to epithelial cells of the testis and epididymis (Korah et al, 2003a, b).

Proper maturation of sperm (ie, acquisition of motility and fertility) occurs as these cells transit through the proximal region of the epididymis, and this is accomplished by the segment-specific activities of the epithelial cells lining this duct (Orgebin-Crist and Olson, 1984; Cornwall et al, 2002; Robaire et al, 2006). The epididymis in many mammals is composed of principal, clear, basal, narrow, and apical cells, all of which contribute to fine-tuning the luminal environment to create the proper environment necessary for sperm maturation (Hamilton, 1975; Orgebin-Crist et al, 1975). These epithelial cells monitor proper water balance, ionic composition, and pH of the lumen and perform activities such as secretion and endocytosis of proteins and protection and transportation of sperm (Robaire and Hermo, 1988; Hermo and Robaire, 2002; Breton, 2003). The number of epididymal proteins implicated in these various processes continues to grow, which attests to the marked diversity and complexity of this tissue (Robaire et al, 2006). Thus, alterations in the expression of proteins of these cellular activities will have an effect on the structure and functions of the epididymal epithelium, which could impact sperm maturation.

The epididymis of PPCA –/– mice develops major structural alterations in principal, clear, narrow, and basal cells of the epithelium with the caput and corpus regions being the most dramatically affected (Korah et al, 2003b). Notable are the accumulation of pale lysosomes within the cytoplasm of these cells, often overwhelming the other organelles. Quantitative analyses have confirmed that outer profile areas of epididymal tubules in caput and corpus regions of PPCA –/– mice are increased in size compared with controls. This is associated with a significant increase in the size of the epithelium with a concomitant decrease in luminal profile area, both suggestive of a swollen epithelial lining (Korah et al, 2003b). Sperm are found in the epididymal lumen; PPCA –/– mice are fertile, but their fertility status, based on preliminary counts of pups per litter, appears to decline with age (d'Azzo, unpublished data). Another prominent feature of PPCA –/– mice is the accumulation of macrophages within the intertubular spaces of the epididymis. These cells are engorged with pale lysosomes, reflecting the absence of PPCA in their lysosomes (Korah et al, 2003b). The presence of these cells, as well as intraepithelial halo cells that are also grossly altered, suggests that antigens may be leaking from the epididymal lumen as a result of improper functioning of the blood-epididymal barrier.

The blood-epididymal barrier is formed by the presence of a complex number of apical tight junctions and adherens junctions between adjacent epithelial cells (Cyr et al, 2002). Claudins (Cldns) are a large family of at least 24 transmembrane proteins which have been shown to be essential for the structural integrity of tight junctions (Furuse et al, 1998; Morita et al, 1999; Sonoda et al, 1999; Tsukita and Furuse, 2000; Guan et al, 2005). Recent studies have shown that occludin and Cldns are prominent components of tight junctions between epididymal epithelial cells (Cyr et al, 1999; Gregory et al, 2001; Guan et al, 2005; Gregory and Cyr, 2006).

In the present study, the effects of PPCA deficiency on the integrity of the blood-epididymal barrier and functions of the epididymal epithelium were assessed using lanthanum nitrate as a tracer to monitor the barrier integrity and light microscopy (LM) immunocytochemistry to characterize differences in the expression of 4 members of the Cldn family of tight junction—sealing proteins. Part of the effectiveness of epididymal functions was evaluated by indirectly characterizing the motility characteristics of sperm taken from the cauda epididymides of PPCA-deficient mice.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgement
  7. References

Animals

A mouse model of galactosialidosis (PPCA –/–) was developed through targeted disruption of the PPCA gene. A targeting vector for homologous recombination was constructed, which gave rise to a null mutation at the PPCA locus (Zhou et al, 1995). Wild-type and PPCA –/– mice of strain C57Bl6/129Sv at different ages were used for this study. All experimentation with these mice was done in strict accordance with guidelines defined by the Animal Care Committees of McGill University and the St Jude Children's Research Hospital.

Tracer Studies

Three PPCA +/+ (wild type) and 3 PPCA –/– mice at 7 months were anesthetized with single intraperitoneal injections of sodium pentobarbital (0.01 mL/10 g; MTC Pharmaceuticals, Hamilton, Canada). Testes and epididymides were perfused via the vascular system using several solutions introduced in a sequential manner through the heart: 1) Ringer lactate solution for 1 minute to remove the blood, 2) 1% lanthanum hydroxide in Ringer lactate solution for 2 minutes as an intercellular tracer, and 3) 1% lanthanum hydroxide and 2.5% glutaraldehyde in 0.1 M sodium cacodylate buffer containing 0.05% CaCl2 for 10 minutes as a precipitate-preserving fixative. After perfusion, the epididymides were removed and divided into the initial segment, caput, corpus, and cauda (Hermo et al, 1991a). Tissues from each region of the epididymis were subsequently cut into 1-mm3 pieces and fixed overnight at 4°C in 0.1 M sodium cacodylate buffer (without lanthanum) containing 2.5% glutaraldehyde. On the following day, the tissues were washed 3 times in 0.1 M sodium cacodylate buffer containing 0.2 M sucrose and incubated in buffer overnight at 4°C. The tissues were postfixed in 1.5% potassium ferrocyanide + 1% osmium tetroxide for 1 hour at 4°C (Karnovsky, 1971) and subsequently processed for routine electron microscopic analyses, as described previously (Hermo et al, 1991b). Thin sections were cut with a diamond knife and stained with uranyl acetate and lead citrate. All electron micrographs were taken with a FEI Tecnai-12 electron microscope (FEI Company, Hillsboro, Ore).

Immunolocalization

PPCA +/+ and PPCA –/– mice at 3 and 7 months (2–3 mice per age) were anesthetized with sodium pentobarbital and fixed by cardiac perfusion with Bouin fixative. In addition, 2 PPCA +/+ and 2 PPCA –/– mice at 7 months were anesthetized and fixed by cardiac perfusion with St Marie fixative (Cyr et al, 1999). Following perfusions, the epididymides were removed (both sides) and partitioned into their 4 major regions, as described above. The tissues were immersed in fresh fixative for 72 hours and then dehydrated and embedded in paraffin. Tissue sections at 5-μm thickness were cut and mounted on glass slides.

Primary antibodies used in this study (summarized in Table 1) included: 1) rabbit anti—Cldn-1 (51–9000; Zymed Laboratories, South San Francisco, Calif; 1:50 dilution), 2) rabbit anti—Cldn-3 (34–1700; Zymed Laboratories; 1:50 dilution), 3) rabbit anti—Cldn-8 (GTX77832; Gene Tex Inc, San Antonio, Tex; 1:100 dilution), and 4) rabbit anti—Cldn-10 (ab24792; Abcam Inc, Cambridge, Mass; 1:100 dilution). Dilution buffer was purchased from DakoCytomation (S0809; Mississauga, Canada). Sections of Bouin- and St Marie—fixed epididymides were deparaffinized with Histoclear (Fisher Scientific Company, Ottawa, Canada) and rehydrated in a series of 100%, 100%, 95%, 80%, 70%, and 50% ethanol solutions, 0.3 M glycine, and 1 × phosphate-buffered saline. Antigen retrieval was accomplished by boiling the sections in a microwave at full power for 2 to 3 minutes in 0.1 M citrate buffer (pH 6.0), followed by heating for 7 minutes at 60% power of the microwave. Endogenous peroxidase activity was blocked for 5 to 10 minutes with a peroxidase-blocking reagent (DakoCytomation), followed by washing. Immunolocalization of Cldns was performed using the Envision+ peroxidase diaminobenzidine (DAB) kit (K4010; DakoCytomation). Washings between each step were conducted for 10 minutes using a buffer solution containing 0.05 M Tris, 0.3 M NaCl, and 0.1% Tween 20 (pH 7.2–7.6). Substrate-chromogen solutions were prepared by adding 24 μL of liquid DAB + chromogen to 1 mL of the substrate buffer. The sections were counterstained for 10 seconds in a 1:5 diluted solution of 0.1% methylene blue and 0.1% thionin, washed, and quickly dehydrated with Histoclear. Coverslips were mounted on slides with Permount. Controls consisted of treating the sections with normal rabbit serum at a dilution similar to that of the primary antibody. All digital images were taken with an Infinity USB 2.0 Hi-Speed camera (Lumenera Scientific, Ottawa, Canada).

Table 1. . Relative intensities of claudin expression in principal cells of different regions of the epididymides of PPCA +/+ and PPCA -/- mice*†
  
  ALBALBALBALB
  1. *Data based on 3 mice per group (4 slides/animal) employing light microscopy immunocytochemistry on a minimum of 3 runs for each antibody

  2. †A indicates apical; B, basal; C, reaction becomes cytoplasmic; L, lateral; PPCA, protective protein cathepsin A; +, moderate reaction; ++, intense reaction; +/–, reaction appears over some but not all cells; and –, absence of reaction

Claudin 1PPCA +/++/–++++++++
 PPCA –/–++C+C+++
Claudin 3PPCA +/++/–+/–++++++++
 PPCA –/–C++++++
Claudin 8PPCA +/++/–+++++++++/–
 PPCA –/–+/–+/–+/–++/–+++/–
Claudin 10PPCA +/++/–+++
 PPCA –/–+/–+

Sperm Collection for Motility Analyses

Four PPCA +/+ and 3 PPCA –/– mice at 7 months were weighed and anesthetized with isoflurane. The cauda region of the left epididymis of each mouse was dissected and placed in a freezer at −20°C for subsequent sperm count analyses. The right cauda epididymis of each animal was clamped both proximally and distally and then excised and rinsed in a 35-mm plastic Petri dish containing prewarmed Hanks medium M199 (Invitrogen Canada Inc, Burlington, Canada) supplemented with 0.5% bovine serum albumin at 37°C. It was then transferred to a fresh Petri dish and the cauda region was pierced using a surgical blade (number 11; Fisher Scientific, Nepean, Canada), allowing sperm to disperse into the medium. After removal of the cauda tissue, the Petri dish was placed in an incubator at 37°C in a 5% CO2 atmosphere for 5 minutes. Subsequently, an aliquot of the sperm suspension was placed into an 80-μm glass chamber and analyzed using an IVOS automated semen analyzer (Hamilton Thorne Biosciences, Beverly, Mass) as designated for mouse samples. Correlation and statistical analyses of data and power tests were conducted with the Statistica Data Miner (version 7.1; Statsoft Inc, Tulsa, Okla). For univariate factorial analysis of variance test and post hoc unequal N HSD t-tests (continuous variables) and Fisher's exact tests (variables as proportions), P < .05 was considered significant. Differences between the sums of correlation coefficients were computed as: (ΣPearson r for parameter A across parameters A—N in PPCA –/– mice) — (ΣPearson r for parameter A across parameters A—N in PPCA +/+ control mice). Principal component analyses of covariances were computed with scaled data using the multivariate statistical package (version 3.1; Kovach Computing Services, Pentraeth, United Kingdom).

Sperm Counts

The frozen left cauda epididymidis of each animal was thawed and homogenized in a 50-mL conical tube containing 1 mL of distilled water. The tissue was then homogenized on ice for 1 to 2 minutes with a Polytron. A 100-μL aliquot of the resulting homogenate was placed in a 1.5-mL microcentrifuge tube coated with IDENT fluorescent dye (Hamilton Thorne Biosciences) and incubated for 5 to 10 minutes at room temperature. The solution was mixed, and a 5-μL aliquot was placed in a 20-μm sperm analysis chamber (2 Cel; Hamilton Thorne Biosciences) and quantified with the IVOS semen analyzer under ultraviolet light using the IDENT option/static sperm. Log10 transformations were done prior to carrying out t-tests assuming unequal variances; P < .05 was considered significant.

Results

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgement
  7. References

Lanthanum nitrate was used to test the patency of the blood-epididymal barrier. This tracer was found within the intertubular spaces between adjacent epididymal tubules as well as myoid cells enveloping the tubules in both PPCA +/+ and PPCA –/– mice (not shown). The tracer filled the intercellular spaces, extending to the level of the apical tight junctional complexes between adjacent epithelial cells (Figure 1a and c). The tracer did not penetrate beyond the junctional complexes in PPCA +/+ mice. However, in PPCA –/– mice, the tracer extended through the tight junctional complex to the very edge of the lumen (Figure 1b and d) and was present in approximately 40% of epididymal lumens of PPCA −/− mice (Figure 1b and inset).

image

Figure 1. . Electron microscope images of adult wild-type (a and c) and deficient in protective protein cathepsin A (b and d) mice perfused with lanthanum nitrate as an intercellular tracer. In a and c, lanthanum nitrate, seen as a dense precipitate, percolates the intercellular spaces (arrowheads) between adjacent principal cells up to the level of the apical junctional complexes. The latter (arrows) are not infiltrated with lanthanum nitrate, suggesting that the tight junctions prevent entry into these areas. There is no evidence of lanthanum nitrate in the epididymal lumen. In b and d, lanthanum nitrate permeates the entire intercellular space (arrowheads) including the apical junctional complexes (arrows). Lanthanum nitrate is also evident in the epididymal lumen (circles). In the inset of b, lanthanum nitrate (circles) is seen freely distributed in the epididymal lumen adjacent to sperm heads. Lu indicates lumen; Mv, microvilli; Ly, lysosomes; S, sperm head; m, mitochondrion; G, Golgi apparatus; and F, sperm flagellum. Original magnification, 8200× (a, b, and d); 9900× (c).

Download figure to PowerPoint

Immunolocalization of Cldns 1, 3, 8, and 10 indicated that while their expression in PPCA –/– mice was not abolished, dramatic changes in their localization were evident. Immunoreactions were defined as apical, lateral, basal, and cytoplasmic (Table 1). Apical membrane reactions typical of PPCA +/+ mice in all regions (Table 1; Figure 2a, c, e, and f) were diminished or focalized to apical cytoplasmic reactions in most regions in PPCA –/– mice (Table 1; Figure 2b, d, f, and h) in the case of Cldn 1. Cldn 3 exhibited apical and lateral immunoreactions between adjacent principal cells in the initial segment and caput regions of PPCA +/+ mice (Figure 3a and c). The immunoreactions were decreased in the initial segment of PPCA −/− mice (Figure 3b), and Cldn 3 was localized to the cytoplasm of principal cells of the caput region (Table 1; Figure 3d). Other regions showed no differences between PPCA +/+ and PPCA −/− mice (Table 1; Figure 3e and f). Apical and basal membrane immunoreactions of Cldn 8 in PPCA +/+ mice were reduced in the caput and corpus regions of PPCA −/− mice (Table 1; Figure 4a through f). For Cldn 10 (Figure 5a through f), the caput and corpus regions showed no apical membrane staining in PPCA −/− mice (Figure 5b and d compared with Figure 5a and c). However, similar immunoreactions were noted in the cauda region between wild-type and PPCA −/− mice (Figure 5a and c compared with Figure 5e and f). Negative control sections incubated without primary antibody showed no immunoreaction anywhere within the epididymis (Figure 4g and h).

image

Figure 2. . Initial segment (a and b), caput (c and d), corpus (e and f), and cauda (g and h) regions of the epididymis of wild-type (a, c, e, and g) and deficient in protective protein cathepsin A (PPCA -/-; b, d, f, and h) mice stained with an anti—Cldn-1 antibody. Major alterations between wild-type and PPCA -/- mice include a reduction of staining in the apical reaction (large arrows) of the initial segment, complemented by the presence of a cytoplasmic reaction in the caput and corpus regions. Lateral (arrowheads) and basal (small arrows) reactions do not show altered phenotypes between wild-type and PPCA -/- mice. Lu indicates lumen; IT, intertubular space; stars, vacuoles in cytoplasm; and C, clear cells. Original magnification, 100×.

Download figure to PowerPoint

image

Figure 3. . Initial segment (a and b), caput (c and d), and corpus (e and f) regions of the epididymis of wild-type (a, c, and e) and deficient in protective protein cathepsin A (PPCA -/-; b, d, and f) mice immunostained with an anti—Cldn-3 antibody. Major alterations between wild-type and PPCA -/- mice include reduced apical (large arrows) and lateral (arrowheads) staining in the initial segment and caput region, complemented by an apical cytoplasmic reaction in the caput epididymidis. Lu indicates lumen; IT, intertubular space; P, principal cells; and C, clear cells. Original magnification, 250×.

Download figure to PowerPoint

image

Figure 4. . Caput (a and b), corpus (c and d), and cauda (e and f) regions of wild-type (a, c, and e) and deficient in protective protein cathepsin A (PPCA -/-; b, d, and f) mice immunostained with an anti—Cldn-8 antibody. Reduced apical (large arrows) and basal (small arrows) reactions are seen in the caput and corpus regions. When epididymal tissues of wild-type (g) and PPCA -/- (h) mice were stained in the absence of primary antibody, there was a complete absence of reaction throughout this tissue. Lu indicates lumen; IT intertubular space; P, principal cells; and C, clear cells. Original magnification, 100×.

Download figure to PowerPoint

image

Figure 5. . Initial segment (a and b), caput (c and d), and cauda (e and f) regions of wild-type (a, c, and e) and deficient in protective protein cathepsin A (PPCA -/-; b, d, and f) mice immunostained with an anti—Cldn-10 antibody. There is a reduction of apical staining (large arrows) in the caput and corpus regions. Note that there is a reaction (circles) around the border of the nucleus of wild-type and PPCA -/- mice in the initial segment and caput regions. Lu indicates lumen; IT, intertubular space; and P, principal cells. Original magnification, 250×.

Download figure to PowerPoint

Sperm concentrations in the cauda epididymis were dramatically different. In PPCA –/– mice, sperm counts were 70% lower than in PPCA +/+ mice (Table 2). Five parameters defining sperm motility behavior, including percent motile, progressively motile, and rapid sperm and the velocity descriptors smooth-path velocity (VAP) and straight-line velocity (VSL), were also significantly reduced (Table 2). In contrast, the sperm descriptor amplitude of lateral head displacement (ALH) was mildly, but significantly, greater in PPCA –/– mice (Table 2). Parameters beat-cross frequency (BCF) and percent static sperm showed large significant increases by 144% and 314%, respectively, in PPCA –/– mice (Table 2). Relationships within the medium and slow categories of sperm movement were unclear; however, additional analyses revealed that the percentage of medium- and slow-moving sperm were negatively correlated to percent static sperm in the most severe case (Table 2; Figure 6).

Table 2. . Sperm counts and motility changes comparing PPCA −/−with PPCA +/+ mice*†
ParameterPPCA +/+ mice Mean ± SD (# of observations)PPCA –/– mice Mean ± SD (# of observations)Change‡P§ Power‖
  1. *ALH, amplitude of lateral head displacement (μm); BCF, beat-cross frequency (Hz); LIN, linearity (ratio of VSL:VCL); NS, not significant; PPCA, protective protein cathepsin A; STR, straightness (ratio of VSL:VAP); VAP, smooth-path velocity (μm/sec); VCL, track velocity (μm/sec); and VSL, straight-line velocity (μm/sec)

  2. †Total number of observations (measurements) made on a pool of 4 PPCA +/+ and 3 PPCA –/– mice

  3. ‡PPCA –/– mice compared with PPCA +/+ mice

  4. §P < .05 considered significantly different; Fisher's exact test was used to compare differences between means for ratios and percentages

  5. ‖This is the power associated with rejecting the null hypothesis that the 2 means are equal. The Z-test for comparing 2 proportions was used in power calculations for variables listed under ratios and percentages

  6. ¶Head shape (ratio of minor to major axes of sperm head)

  7. #The number of observations was too low to establish any significance for these differences

Cauda epididymis      
Sperm counts, millions/mL23.3 ± 4.8 (147)7.1 ± 3.2 (116)−70%.0000 1.0000
For all cases in data set      
Raw values(102)(58)    
VAP61.2 ± 1.552.5 ± 19.1−14%.0010 .9454
VSL5.1 ± 9.341.6 ± 16.3−17%.0002 .9762
VCL87.8 ± 15.686.5 ± 29.4−1%.7467NS.0653
ALH2.5 ± .63.2 ± 2.028%.0046 .8711
BCF.9 ± 1.72.2 ± 4.7144%.0303 .6679
Ratios      
STR81.0 ± 9.277.8 ± 7.1−4%.6283NS.0534
LIN58.8 ± 7.249.8 ± 9.9−15%.2724NS.1545
Elong(ation)¶46.2 ± 4.85.1 ± 7.98%.6356NS.0544
Percentages      
Motile85.3 ± 7.345.0 ± 24.5−47%.0000 .9995
Prog(ressive)33.6 ± 12.313.1 ± 12.3−61%.0052 .7836
Rapid5.2 ± 13.62.3 ± 16.9−60%.0003 .9612
Medium35.2 ± 12.924.6 ± 15.1−30%.1667NS#.2192
Slow2.7 ± 4.15.4 ± 6.9100%.3844NS#.0799
Static12.0 ± 7.049.7 ± 25.8314%.0000 .9989
    
Raw valuessame as above(42)    
VAP 51.8 ± 2.8−15%.0025 .9294
VSL 41.0 ± 17.8−18%.0007 .9661
VCL 88.5 ± 33.21%.8923NS.0534
ALH 3.4 ± 2.336%.0017 .9421
BCF 2.7 ± 5.4200%.0115 .8423
Ratios      
STR 77.0 ± 7.9−5%.6824NS.0589
LIN 47.7 ± 1.0−19%.2251NS.1778
Elong(ation)¶ 51.1 ± 8.411%.5858NS.0576
Percentages      
Motile 33.5 ± 16.8−61%.0000 1.0000
Prog(ressive) 9.0 ± 1.1−73%.0025 .8643
Rapid 14.1 ± 13.2−72%.0001 .9894
Medium 19.4 ± 13.3−45%.0597NS#.3760
Slow 4.7 ± 5.974%.5582NS#.0500
Static 61.8 ± 18.9415%.0000 1.0000
image

Figure 6. . Scatter plots of motility parameters (A) percent static vs percent medium sperm and (B) percent static vs percent slow sperm. Linear regression lines in both panels are computed from all cases (solid lines) or using only those cases in which percent static sperm was >30% (dashed lines). Note in both panels that some motile sperm of deficient protective protein cathepsin A (PPCA -/-) mice (solid symbols) show a range of values that overlaps those observed in PPCA +/+ mice (open symbols). Note also in B the high frequency of cases in which percent slow sperm = 0 for levels of percent static sperm <30% in both PPCA +/+ and PPCA -/- mice. Both graphs imply that there is a trend for the frequency of medium- and slow-moving sperm to decline as the number of static sperm in a sample increases above 30% static cells.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgement
  7. References

In PPCA –/– mice, the marked increase in the number of macrophages within the epididymal intertubular spaces and presence of abnormal halo cells within the epithelium (Korah et al, 2003b) are both suggestive of blood cell responses to antigens leaking in from the epididymal lumen, possibly due to disruption of the blood-epididymal barrier. This barrier, as in other epithelial cells that seal off luminal spaces, is created by a multitude of apically positioned junctional proteins that effectively bind and seal together adjacent epithelial cells, thereby allowing distinct differences in the ionic and molecular compositions of fluid to be maintained in the epididymal lumen compared with the underlying connective tissue spaces and blood vessels (Cyr et al, 2002). This idea has been reinforced in 2 ways in the present study. First, lanthanum nitrate, a well-known intercellular ionic tracer, entered and bypassed the apical junctional complexes in PPCA –/– mice and penetrated freely to the level of the epididymal lumen in some tubules. Second, immunostaining patterns for members of the Cldn family of tight junction proteins were noticeably different between PPCA –/– mice and their wild-type counterparts (Table 1).

Movement of material across cell membranes occurs in an energy-dependent manner through a large number of cell-specific pumps, channels, and transporters. Recent studies have shown that tight junctions regulate solute movement through the paracellular pathway across epithelia (van Itallie and Anderson, 2006). Paracellular barriers vary among epithelia and behave as if they are lined with pores that have charge and size selectivity. Cldns are transmembrane proteins that are essential for the structural integrity of tight junctions (Furuse et al, 1998; Sonoda et al, 1999; Morita et al, 1999; Tsukita and Furuse, 2000; Guan et al, 2005). They are intercellular adhesion molecules that have variable pore-like properties (van Itallie and Anderson, 2006). This appears to be accomplished by the presence of charged amino acids located in the extracellular loops of several Cldns that establish paracellular charge selectivity (Colegio et al, 2002; van Itallie et al, 2003). The presence of Cldns 1 and 2 has been proposed to constitute aqueous pores with high conductance (Furuse et al, 2001). Cldn 10 appears to form anion-selective channels in tight junctions of the rat epididymis (Guan et al, 2005). In the present study, the altered states of several Cldns in PPCA –/– mice suggest that the ion composition of the epididymal lumen may be altered and in this way adversely affect sperm maturation as reflected by altered motility and sperm shape parameters. Indeed, it has been shown that altering the epididymal lumen pH can affect sperm motility (Morton et al, 1974; Turner and Howards, 1978; Pholpramool and Chaturapanich, 1979; Wong et al, 1981). Also, changes in sperm cell volume have been reported for c- Ros—deficient mice that appear to affect several osmolyte transporters (Cooper et al, 2003).

Cldns have been reported in rat epididymis where they show both regional specificity and preferences for either apical, lateral, or basolateral locations on epididymal epithelial cell membranes (Gregory et al, 2001; Guan et al, 2005; Gregory and Cyr, 2006). In the present study, we clearly showed that each of the 4 Cldn family members investigated, which had not previously been localized in mice, showed the same varied immunostaining patterns in principal cells depending on the epididymal region and differences in specific Cldns similar to what was previously described for rat. Apical reactions for Cldns 1 and 3, however, showed plasma membrane immunolocalization in the mouse that were different from those reported for rat (Gregory et al, 2001; Gregory and Cyr, 2006). This suggests that there may be some species variability in proteins associated with junctional complexes in the epididymis between rats and mice.

The distributions of Cldns were dramatically different in PPCA –/– mice compared with wild-type mice (Table 1; Figures 2, 3, 4, 5). In some cases, there was a reduction or complete absence of immunostaining along the apical plasma membranes, where the blood barrier resides, complemented by diffuse apical cytoplasmic reactions. Other plasma membrane domains were only affected in certain epididymal regions and differed depending upon a given Cldn. The presence of apical cytoplasmic reactions suggests that targeting of the respective tight junction protein is compromised in PPCA –/– mice. Apical reactions have been documented for several Cldns during postnatal epididymal development (Guan et al, 2005; Gregory and Cyr, 2006). In the present study, approximately 40% of the tubules were noted to be compromised by administration of lanthanum nitrate, and while we did not perform quantitative analyses, most of these changes were noted in more proximal regions of the epididymis. This suggests that permeability of lanthanum nitrate in the cauda region may be different in PPCA –/– mice and may explain the difference noted for Cldn staining in this region. Indeed, in PPCA –/– mice, the cauda epididymis was noted to be the least affected region of the epididymis in terms of structural abnormalities (Korah et al, 2003b). Whereas we do not know the precise mechanism underlying the link between PPCA −/− deficiency and the altered blood-epididymal barrier, we suggest that the accumulation of lysosomes in both number and size in epithelial cells, which at times completely fills their entire cytoplasm, could compromise other organelles such as the endoplasmic reticulum and Golgi apparatus. This compromise would lead to a decrease in the synthesis of proteins, including those involved in junctional complexes.

In the proximal epididymal regions of most mammals, including the mouse, sperm undergo maturation by acquiring motility and the ability to fertilize the ovum (Orgebin-Crist and Olson, 1984). The acquisition of sperm motility involves quantitative increases in the percentage of motile sperm as well as qualitative differences in their motility behavior (Blandau and Rumery, 1964; Hinton et al, 1979; Cornwall et al, 1986; Soler et al, 1994). The epithelial cells of the proximal epididymis fine-tune the luminal environment in which sperm mature by providing the proper protein and lipid components that interact with the sperm surface to render them mature and by creating the proper pH and ionic and water balance (Cyr et al, 2002). Thus, the integrity of the structure and functions of these cells is an essential prerequisite for proper sperm maturation. The compromised nature of principal cells within the caput and corpus regions of PPCA –/– mice as indicated by leakiness to lanthanum nitrate and altered expression of the 4 Cldns suggests that cells in these regions have diminished functions; this is likely to be reflected as incomplete or improper maturation of sperm. Indeed, numerous macrophages were noted in the epididymal intertubular spaces, suggesting a response to antigens emanating from the lumen, some of which could be derived from sperm. The former may secrete factors in response to these antigens that could enter the lumen, thereby affecting the sperm themselves and their motility behavior.

The results from the motility analyses (Table 2) fully support the idea that sperm maturation is compromised, as indicated by the extreme shift from the motile (and progressively motile) category to the static category for the much reduced numbers of sperm that actually are produced in the PPCA –/– mice. It should be noted that sperm emanating from the testis do not show any signs of structural abnormalities and appear to be comparable to controls, as seen in the electron microscope (Korah et al, 2003a). The present results further show unequivocally that the shift of motility involves sperm from primarily the rapid category and likely to a lesser extent, based on Figure 6, the medium and slow categories in the more severely affected mice. It is also not surprising that differences within the percent medium and slow categories, despite their large differences in relative percent change (−30% vs 100%, respectively), could not be resolved as significant. Indeed, the absolute differences they represent are actually quite small (−11% and 2.7% of total sperm in PPCA –/– mice), and as ratio values, they would therefore require large numbers of observations in controls and especially the experimental groups to be established as significant.

Sperm velocity parameters VAP and VSL were modestly, albeit significantly, depressed, whereas the parameter track velocity (VCL) was unchanged in PPCA –/– mice (Table 2). VCL represents the average velocity over each individual step that a sperm makes while moving. The fact that it remains equal to normal while the velocity calculated between the starting and ending points of the sperm track (VSL) and the velocity computed as an average over all steps in a sperm track (VAP) are shorter suggest that sperm in PPCA –/– mice exert more effort moving side to side than moving forward in space. This conclusion is supported by the findings of a 61% reduction in sperm progressive velocity and much higher BCF and ALH values in the PPCA –/– mice (more center path crossing and thrashing of sperm heads side to side; Table 2). These are all symptomatic of less vigorous sperm being produced in PPCA –/– mice.

Summed correlation difference plots identified 5 motility parameters that appear to stand out more prominently from other parameters in PPCA –/– mice (Figure 7A). These were elongation, BCF, and percent medium, slow, and static sperm (Figure 7A). Elongation, BCF, and percent static sperm in this plot showed increasing magnitudes in mean differences to wild-type mice, but they all shared a similar level of summed correlation differences that is decidedly more negative than the majority of other motility parameters (Figure 7A). Percent slow sperm differed from the majority of motility parameters only as a result of its large relative increase in mean difference from wild-type mice, whereas percent medium sperm showed a mean difference that is slightly negative but within range of other motility parameters and a summed correlation difference that is more positive compared with all other parameters. The same 5 motility parameters were also found to be prominent and arrayed graphically in similar, although not identical, fashion in knockout mice lacking estrogen receptor α (αERKO) (Ruz et al, 2006). These mice are infertile, and the root cause of their problem appears to result from excessive water retention associated with defective expression of transporters in the efferent ducts (Hess, 2002). As documented in this study, PPCA –/– mice have altered expression of Cldns in the caput and corpus regions, making the lumens leaky and upsetting the water and ion balance, which may resemble some of the problems encountered in the αERKO mice.

image

Figure 7. . Scatter plots summarizing (A) changes in the motility behavior of sperm of deficient protective protein cathepsin A (PPCA -/-) mice compared with PPCA +/+ mice and showing variable loadings (arrows as vectors radiating from the graph origin) and (B) case scores (symbols) following principal component analyses of covariances of motility parameters in PPCA -/- and PPCA +/+ mice. (A) Nine motility parameters are relatively tightly grouped together, reflecting their slightly positive or negative changes in mean values or summed intervariable correlations. Percent static shows the greatest change in mean values, whereas elongation has the most negative and percent medium has the most positive summed correlation differences when comparing PPCA −/− with PPCA +/+ mice. (B) The first principal component axis (axis 1, horizontal) accounts for 61% of total variance and defines sperm primarily in primarily in terms of their motile (positive) or static (negative) behavior (vectors projecting from the origin). The second principal component axis (axis 2, vertical) accounts for 19% of total variance and reflects velocity of sperm movement (straight-line velocity, average path velocity, curvilinear velocity; positive) vs sperm moving at medium speed (medium; negative). Case scores plotted on the graph (symbols) clearly illustrate the tendency for sperm samples from PPCA −/− mice (knockout) to contain a few or many static cells as opposed to samples from PPCA +/+ mice (wild type) that contain mostly motile cells. Note that a fraction of sperm from PPCA −/− mice have overall motility characteristics closely resembling normal sperm (areas in which up and down triangles are superimposed).

Download figure to PowerPoint

The unexpected prominence of percent medium sperm within the data set was also confirmed by principal component analyses of covariances in which variable loadings on the first 2 main axes indicated that percent medium sperm forms a strong main component along the negative y-axis opposite the velocity parameters VCL, VSL, and VAP (Figure 7B). Case loadings done on the same principal components plot show clearly the dispersed trend for more static sperm in PPCA –/– mice and the tight trend for motile, progressive, and rapid sperm in normal PPCA +/+ mice (Figure 7B). This plot clearly shows that the velocities of sperm in PPCA –/– and PPCA +/+ mice (VCL, VSL, and VAP) are similar and independent of the number of static sperm in a given sample (Figure 7B, vertical spread of all points plotted). Hence, sperm in PPCA −/− are either static or motile, and if they are motile, then they travel with velocity features resembling, although not exactly the same, as normal mice.

Taken together, results from this study suggest that there are several cumulative root causes for fertility problems in PPCA –/– mice. First, PPCA –/– mice produced 70% fewer sperm compared with normal male mice. Second, within this diminished sperm population, only 45% of sperm were motile; of these, only 20% moved in a rapid manner. In comparative terms, this represents the equivalent of only 6% of sperm moving rapidly in PPCA –/– mice (30% production × 20% rapid = 6%) relative to the 50% of rapidly moving sperm routinely produced by wild-type mice (Table 2). Finally, motile sperm in PPCA –/– mice showed more side to side head movements and path crossings as compared with normal mice, suggesting that these sperm are less forwardly vigorous.

Acknowledgement

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgement
  7. References

We thank Jeannie Mui for her excellent electron microscope technical assistance.

References

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgement
  7. References
  • Blandau RJ, Rumery RE. The relationship of swimming movements of epididymal spermatozoa to their fertilizing capacity. Fertil Steril. 1964;15: 571579.
  • Breton S.. Luminal acidification in the epididymis and vas deferens. In: Hinton BT, Turner TT, eds. Third International Conference on the Epididymis. Charlottesville, Va: Van Doren; 2003: 6072.
  • Colegio OR, van Itallie CM, McCrea HJ, Rahner C., Anderson JM. Claudins create charge-selective channels in the paracellular pathway between epithelial cells. Am J Physiol Cell Physiol. 2002;283: C142C147.
  • Cooper TG, Wagenfeld A., Cornwall GA, Hsia N., Chu ST, Orgebin-Crist M-C, Drevet J., Vernet P., Avram C., Nieschlag E., Yeung CH. Gene and protein expression in the epididymis of infertile c-ros receptor tyrosine kinase-deficient mice. Biol Reprod. 2003;69: 17501762.
  • Cornwall GA, Lareyre J-J, Matusik RJ, Hinton BT, Orgebin-Crist M-C. Gene expression and epididymal function. In: Robaire B., Hinton BT, eds. The Epididymis, From Molecules to Clinical Practice. New York, NY: Kluwer Academic/Plenum Publishers; 2002: 169199.
  • Cornwall GA, Smyth TB, Vindivich D., Harter C., Robinson J., Chang TSK. Induction and enhancement of progressive motility in hamster epididymis. Biol Reprod. 1986;35: 10651074.
  • Cyr DG, Finnson K., Dugresne J., Gregory M.. Cellular interactions and the blood epididymal barrier. In: Robaire B., Hinton BT, eds. The Epididymis, From Molecules to Clinical Practice. New York, NY: Kluwer Academic/Plenum Publishers; 2002: 103118.
  • Cyr DG, Hermo L., Egenberger N., Mertineit C., Trasler J., Laird D.. Cellular immunolocalization of occludin during embryonic and postnatal development of the mouse testis and epididymis. Endocrinology. 1999;140: 38153825.
  • D'Azzo A., Andria G., Strisciuglio P., Galjaard H.. Galactosialidosis. In: Scriver C., Beaudet A., Sly W., Valle D., eds. The Metabolic and Molecular Bases of Inherited Disease. Vol 2, 7th ed. New York, NY: McGraw-Hill Publishing Co; 1995: 28252837.
  • D'Azzo A., Hoogeveen A., Reuser AJ, Robinson D., Galjaard H.. Molecular defect in combined β-galactosidase and neuraminidase deficiency in man. Proc Natl Acad Sci U S A. 1982;79: 45354539.
  • Furuse M., Fujita K., Hiiragi T., Fujimoto K., Tsukita S.. Claudin-1 and-2: novel integral membrane proteins localizing at tight junctions with no sequence similarity to occludin. J Cell Biol. 1998;141: 15391550.
  • Furuse M., Furuse K., Sasaki H., Tsukita S.. Conversion of zonulae occludentes from tight to leaky strand type by introducing claudin-2 into Madin-Darby canine kidney I cells. J Cell Biol. 2001;153: 263272.
  • Galjart NJ, Morreau H., Willemsen R., Gillemans N., Bonten EJ, D'Azzo A.. Human lysosomal protective protein has cathepsin A-like activity distinct from its protective function. J Biol Chem. 1991;266: 1475414762.
  • Gregory M., Cyr DG. Identification of multiple claudins in the rat epididymis. Mol Reprod Dev. 2006;73: 580588.
  • Gregory M., Dufresne J., Hermo L., Cyr D.. Claudin-1 is not restricted to tight junctions in the rat epididymis. Endocrinology. 2001;142: 854863.
  • Guan X., Inai T., Shibata Y.. Segment-specific expression of tight junction proteins, claudin-2 and −10, in the rat epididymal epithelium. Arch Histol Cytol. 2005;68: 213225.
  • Hamilton DW. Structure and function of the epithelium lining the ductuli efferentes, ductus epididymidis and ductus deferens in the rat. In: Greep RO, Astwood EB, eds. Handbook of Physiology. Vol 5. Washington, DC: American Physiological Society; 1975: 259301.
  • Hanna WL, Turbov JM, Jackman HL, Tan F., Froelich CJ. Dominant chymotrypsin-like esterase activity in human lymphocyte granules is mediated by the serine carboxypeptidase called cathepsin A-like protective. J Immunol. 1994;153: 46634672.
  • Hermo L., Green H., Clermont Y.. Golgi apparatus of epithelial principal cells of the epididymal initial segment of the rat: structure, relationship with endoplasmic reticulum, and role in the formation of secretory vesicles. Anat Rec. 1991b;229: 159176.
  • Hermo L., Robaire B.. Epididymal cell types and their functions. In: Robaire B., Hinton BT, eds. The Epididymis, From Molecules to Clinical Practice. New York, NY: Kluwer Academic/Plenum Publishers; 2002: 81102.
  • Hermo L., Wright J., Oko R., Morales CR. Role of epithelial cells of the male excurrent duct system of the rat in the endocytosis or secretion of sulfated glycoprotein-2 (clusterin). Biol Reprod. 1991a;44: 11131131.
  • Hess RA. The efferent ductules: structure and functions. In: Robaire B., Hinton BT, eds. The Epididymis, From Molecules to Clinical Practice. New York, NY: Kluwer Academic/Plenum Publishers; 2002: 4980.
  • Hinton BT, Dott HM, Setchell BP. Measurement of the motility of rat spermatozoa collected by micropuncture from the testis and from different regions along the epididymis. J Reprod Fertil. 1979;55: 167172.
  • Itoh K., Kase R., Shimmoto M., Satake A., Sakuraba H., Suzuki Y.. Protective protein as an endogenous endothelin degradation enzyme in human tissues. J Biol Chem. 1995;270: 515518.
  • Jackman HL, Morris PW, Deddish PA, Skedgel RA, Erdos EG. Inactivation of endothelin I by deamidase (lysosomal protective protein). J Biol Chem. 1992;267: 28722875.
  • Karnovsky MJ. Use of ferrocyanide-reduced osmium tetroxide in electron microscopy (abstract). In: Proceedings of the 11th American Society of Cell Biology. New Orleans, La: American Society of Cell Biology; 1971;284: 146.
  • Korah N., Smith CE, D'Azzo A., El-Alfy M., Hermo L.. An increase in macrophages in the testis of cathepsin A deficient mice suggest an important role for these cells in the interstitial space of this tissue. Mol Reprod Dev. 2003a;64: 302320.
  • Korah N., Smith CE, D'Azzo A., Mui J., Hermo L.. Characterization of cell- and region-specific abnormalities in the epididymis of cathepsin A deficient mice. Mol Reprod Dev. 2003b;66: 358373.
  • Morita K., Furuse M., Fujimoto K., Tsukita S.. Claudin multigene family encoding four-transmembrane domain protein components of tight junction strands. Proc Natl Acad Sci U S A. 1999;96: 511516.
  • Morton B., Harrigan-Lum J., Albagi L., Jooss T.. The activation of motility in quiescent hamster sperm from the epididymis by calcium and cyclic nucleotides. Biochem Biophys Res Commun. 1974;56: 372379.
  • Orgebin-Crist M-C, Danzo BJ, Davies J.. Endocrine control of the development and maintenance of sperm fertilizing ability in the epididymis. In: Greep RO, Astwood EB, eds. Handbook of Physiology. Vol 5. Washington, DC: American Physiological Society; 1975: 319338.
  • Orgebin-Crist M-C, Olson GE. Epididymal sperm maturation. In: The Male in Farm Animal Reproduction, Courot M. ed., Amsterdam, The Netherlands: Martinus Nijhoff; 1984: 80102.
  • Pholpramool C., Chaturapanich G.. Effect of sodium and potassium concentrations and pH in the maintenance of motility of rabbit and rat epididymal spermatozoa. J Reprod Fertil. 1979;7: 245251.
  • Robaire B., Hermo L.. Efferent ducts, epididymis, and vas deferens: structure, functions, and their regulation. In: Knobil E., Neill JD, eds. The Physiology of Reproduction. New York, NY: Raven Press; 1988: 9991080.
  • Robaire B., Hinton BT, Orgebin-Crist MC. The epididymis. In: Knobil E., Neill JD, eds. Physiology of Reproduction. 3rd ed. New York, NY: Academic Press; 2006: 10711147.
  • Rottier RJ, Hahn CN, Mann LW, Del Martin M Pilar, Smeyne RJ, Suzuki K., D'Azzo A.. Lack of PPCA expression only partially coincides with lysosomal storage in galactosialidosis mice: indirect evidence for spatial requirement of the catalytic rather than the protective function of PPCA. Hum Mol Genet. 1998;7: 17871794.
  • Ruz R., Gregory M., Smith CE, Cyr DG, Lubahn DB, Hess RA, Hermo L.. Expression of aquaporins in the efferent ductules, sperm counts, and sperm motility in estrogen receptor-α deficient mice fed lab chow versus casein. Mol Reprod Dev. 2006;73: 226237.
  • Sohma O., Mizuguchi M., Takashima S., Satake A., Itoh K., Sakuraba H., Suzuki Y., Oyanagi K.. Expression of protective protein in human tissue. Pediatr Neurol. 1999;20: 210214.
  • Soler C., Yeung CH, Cooper TG. Development of sperm motility patterns in the murine epididymis. Int J Androl. 1994;17: 271278.
  • Sonoda N., Furuse M., Sasaki H., Yonemura S., Katahira J., Horiguchi Y., Tsukita S.. Clostridium perfringens enterotoxin fragment removes specific claudins from tight junction strands: evidence for direct involvement of claudins in tight junction barrier. J Cell Biol. 1999;147: 195204.
  • Tsukita S., Furuse M.. Pores in the wall: claudins constitute tight junction strands containing aqueous pores. J Cell Biol. 2000;149: 1316.
  • Turner TT, Howards SS. Factors involved in the initiation of sperm motility. Biol Reprod. 1978;18: 571578.
  • van Der Spoel A., Bonten E., D'Azzo A.. Transport of human lysosomal neuraminidase to mature lysosomes requires protective protein/cathepsin A. EMBO J. 1998;17: 15881597.
  • van Itallie CM, Anderson JM. Claudins and epithelial paracellular transport. Annu Rev Physiol. 2006;68: 403429.
  • van Itallie CM, Fanning AS, Anderson JM. Reversal of charge selectivity in cation or anion-selective epithelial lines by expression of different claudins. Am J Physiol Renal Physiol. 2003;285: F1078F1084.
  • van Pelt J., van Bilsen DG, Kamerling JP, Vliegenthart JF. Structural analysis of O-glycosidic type of sialyloligosaccharide-alditols derived from urinary glycopeptides of a sialidosis patient. Eur J Biochem. 1988;174: 183187.
  • Wong PYD, Lee WM, Tsang AYF. The effects of sodium and amiloride on the motility of caudal epididymal spermatozoa of the rat. Experientia. 1981;37: 6971.
  • Zhou XY, Morreau H., Rottier R., Davis D., Bonten E., Gillemans N., Wenger D., Grosveld FG, Doherty P., Susuki K., Grosveld GC, D'Azzo A.. Mouse model for the lysosomal disorder galactosialidosis and correction of the phenotype with overexpressing erythroid precursor cells. Genes Dev. 1995;9: 26232634.
  • Zhou XY, van Der Spoel A., Rottier R., Hale G., Willemsen R., Berry GT, Strisciuglio P., Andria G., D'Azzo A.. Molecular and biochemical analysis of protective protein/cathepsin A mutations: correlation with clinical severity in galactosialidosis. Hum Mol Genet. 1996;5: 19771987.
Footnotes
  1. This study was supported by grants from the CIHR (LH), NSERC/CIHR Collaborative Health Grant (D.G.C., L.H.), NSERC (D.G.C.), NIHR01DK52025 and GM60905 (A.D.), and the National Institutes of Health (C.E.S.; DE013237)