Although there have been many fascinating studies on cryptdins, the information for each cryptdin isoform was not completely provided. In this study, the postnatal changes in the gene expression of cryptdin 1–6 were evaluated, and the patterns of change were compared between conventional and germ-free mice. Two patterns of postnatal change were observed: gene expression of cryptdins 1, 3 and 6 increased gradually, and that of cryptdins 2 and 5 increased rapidly. Gene expression of cryptdin 4 increased gradually in the ileum but rapidly in the jejunum. Conventional mice showed significantly higher gene expression for all isoforms than germ-free mice. Interestingly, the difference in the gene expression for cryptdin 2, 4 and 5 between the jejunum and ileum seemed to be increased by the presence of the luminal bacteria. The results indicate that cryptdin isoforms develop differently depending on the isoform type, and that the gene expression of all cryptdin isoforms was affected by the presence of the luminal bacteria.
The intestinal tract of monogastric animals is home to an enormous indigenous microbiome. Remarkably, the bacterial load in the small intestinal lumen is c. 104–106-fold lower per gram of contents than in the adjacent colon (Savage, 1977). This is an important difference with regard to the digestion and absorption of nutrients in the small intestine, because digestion and absorption would be less efficient in the presence of a large population of competing bacteria, even if they are not pathogenic. In addition to the indigenous bacteria, a variety of bacteria, including some potential pathogens, are introduced daily into the small intestine through the ingestion of food and drink (Cunliffe, 2003).
A unique lineage of epithelial cells, the Paneth cells, are found in the small intestine of a variety of mammals, but not in the large intestine, and they are thought to play an important role in the maintenance of the environment of the small intestine. The Paneth cells are granulated epithelial cells found at the bottom of the crypts and are derived from intestinal stem cells as well as other epithelial cell lineages such as enterocytes and goblet cells (Staley & Trier, 1965; Trier, 1966; Cheng & Leblond, 1974; Bjerknes & Cheng, 1981; Hermiston et al., 1993). Several proteins, such as lysozyme, secretory phospholipase A2, matrilysin, α1-antitrypsin, tumor necrosis factor-α and α-defensins, are all produced in the Paneth cells and are associated with host defense (Darmoul et al., 1997).
α-Defensins, termed cryptdins in mice, are produced extensively by the Paneth cells and are regarded as an important component of innate immunity in the small intestine (Ouellette & Bevins, 2001). There has been detailed information about the cryptdins from previous studies, and most of it was provided by Professor Andre J. Ouellette and his colleagues (Ouellette et al., 1989, 1994; Selsted et al., 1992; Bry et al., 1994; Huttner et al., 1994; Darmoul & Ouellette, 1996; Darmoul et al., 1997; Ouellette & Bevins, 2001).
According to these studies, cryptdins are cationic peptides of 3–4 kDa and comprise six cysteine residues in a characteristic disulfide bonding motif (Huttner et al., 1994). Until date, six isoforms, cryptdins 1–6, have been purified from the mouse small intestine and all of them possess antimicrobial activity against various bacteria, such as Escherichia coli and Salmonella typhimurium (Ouellette et al., 1994).
Ouellette (1994) indicated that the isoforms have different degrees of antimicrobial activity against E. coli. In an agar diffusion assay, cryptdin 4 seemed to have the strongest antimicrobial activity against E. coli, and in a bacteriocidal assay, cryptdin 2 apparently possessed the strongest antimicrobial activity against E. coli. In addition, a higher level of gene expression for cryptdin 4 was observed in the ileum than in the jejunum (Darmoul & Ouellette, 1996). These facts imply that the different isoforms have different roles in the small intestinal innate immunity and are regulated by different factors.
It is also of interest that the cryptdin genes are expressed even in the small intestine of newborn mice, which do not have the morphologically matured crypts (Ouellette et al., 1989). Predictably, the gene expression levels of cryptdins in newborn mice are lower than those in adults (Ouellette et al., 1989; Darmoul et al., 1997). Similar developmental increase of α-defensin gene expression (HD-5 and HD-6) was also demonstrated in humans (Mallow et al., 1996). Surprisingly, however, it has been suggested that the postnatal increase in cryptdin gene expression seems to be independent of both T-cell involvement and intestinal bacteria (Ouellette et al., 1989). According to Ouellette (1989), gene expression of cryptdins reached the adult level by 4 weeks of age, and subsequently the expression level became similar in conventional and germ-free mice.
However, the probe used for hybridization in the study by Ouellette (1989) had high homology to three cryptdin isoforms (cryptdins 1, 3 and 6). Thus, the postnatal development of each cryptdin isoform and the difference in the expression level of each isoform between conventional and germ-free mice are not yet fully understood. In view of the above suggestion that each cryptdin isoform may have a different role in innate immunity in the small intestine, investigations focusing on each cryptdin isoform are worth conducting.
The aims of this study were to analyze the postnatal changes in the gene expression of each cryptdin isoform (cryptdins 1–6) in the small intestine of the mouse, and to compare the gene expression level of each cryptdin isoform between conventional and germ-free mice at different ages. Because the cDNAs of the cryptdin isoforms, except for cryptdin 4 and 5, have more than 93% nucleotide sequence identity overall (Ouellette et al., 1994), the real-time PCR technique using Taqman probe was used in this study. This is a highly sensitive and specific method and can be used for the analysis of mRNA expression of each cryptdin isoform.
We also evaluated the postnatal timing of the appearance of Paneth cells by light microscopy, because this may be closely related to cryptdin gene expression.
Materials and methods
Animals and sample collection
Fifteen pregnant BALB/c mice were obtained from a commercial supplier (Japan SLC, Shizuoka, Japan) and pups derived from these dams were used in this study. The day when the pups were delivered was regarded as day 0 of postnatal life. The animals were housed under conventional circumstances in our facilities, and the pups were not separated from their dams until weaning at day 21. Three or four mice (mostly four) were sacrificed and dissected on each day from birth to day 25, and days 30, 35 and 40 (maturity) of life (giving a total of 114 mice). One pregnant mouse was dissected 1 day before the scheduled day of delivery, and samples were taken from its fetuses.
The entire small intestine was collected from each mouse and from the fetuses and the intestine was evenly divided into eight segments using a ruler. The second and eighth segments from the pylorus (regarded as the jejunum and ileum, respectively) were used as samples for histochemical evaluations and RNA extraction. Each segment was subdivided into two segments, and the segment on the caecum side was fixed in 10% neutral formalin. The segment on the pylorus side was immediately immersed in RNAlater (Ambion, Tokyo, Japan) and fixed overnight at 4 °C. The samples were then stored at −80 °C until subsequent RNA extractions. Because cryptdins are specific to the Paneth cells in the adult small intestine (Selsted et al., 1992; Ouellette et al., 1994; Selsted & Ouellette, 1995), the isolation of the crypt was omitted in this study to simplify the procedure.
Germ-free BALB/c mice aged 13, 19, 22 and 40 days were purchased from Sankyo Labo Service (Tokyo, Japan); the samples collected were the same as those collected from the conventional mice.
The experimental animals were handled in accordance with the guidelines for animal studies of the Kyoto Institute of Nutrition and Pathology.
RNA extraction and reverse transcription
Total RNA was extracted from the samples using the QuickGene RNA tissue kit SII (Fuji Film, Tokyo, Japan), an RNA extraction kit for use with a semi-automated nuclear acid extraction machine (QuickGene810; Fuji Film).
The concentration of the extracted total RNA was measured with a Nano-Drop ND1000 spectrophotometer (NanoDrop Technologies, Wilmington, DE), and 150 ng of the total RNA was used for reverse transcription using a PrimeScript® RT reagent kit (Perfect Real Time) (Takara, Osaka, Japan). All procedures were performed according to the manufacturer's instructions.
Real-time quantitative PCR
Real-time PCR was performed using an Applied Biosystems 7300 real-time PCR System (Applied Biosystems, Tokyo, Japan). The primers and Taqman probes used in this study are listed in Table 1.
|Genbank accession number||Sequence 5′-3′|
|Genbank accession number||Sequence 5′-3′|
The synthesis of the primers and probes was ordered to Bioresearch Technologies Japan (Tokyo, Japan). The probes were labeled at the 5′ end with FAM and 3′ end with Black Hole Quenchers®.
The primers and probes for cryptdin 3 and 6 can cross-react with the gene for cryptdin 6 and 3, respectively. However, the reactivity of the primers and probes for cryptdin 3 to the target gene is much higher (1000-fold) than to cryptdin 6 gene. In case the primers and probe for cryptdin 6, the reactivity to the target gene is 10-fold higher than to cryptdin 3. Considering the ratio of cryptdin 3 to cryptdin 6 in the mouse small intestine (Ouellette & Bevins, 2001), this cross-reactivity does not significantly affect the results of the study. No effect of this cross-reactivity on the analyses was validated by our preliminary evaluation using the partially cloned cryptdin 3 and 6 genes. The primers and probes for the other four cryptdin genes do not cross-react with the evaluated six cryptdin genes according to the GenBank database.
Amplification was carried out in a 10 µL reaction volume containing 5 µL of Premix Ex Taq™ (Perfect Real Time) (Takara), 0.4 µL cDNA, 0.9 µM of each primer and 0.25 µM of Taqman probe. The thermal cycling profile was 10 s at 95 °C followed by 40 cycles of 5 s at 95 °C and 34 s at 60 °C. The gene for glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was amplified as a housekeeping gene in the same manner as the cryptdin isoforms, but SYBR Premix Ex Taq™ (Perfect Real Time) (Takara) was used instead of Premix Ex Taq™ (Perfect Real Time).
To prepare a standard curve, plasmids containing the PCR product of each gene in pGEM-T easy vector (Promega, Tokyo, Japan) were generated and included in every run. The transcript copy number for each sample was calculated from the standard curve and normalized to the copy number for GAPDH. The data for each cryptdin isoform were calculated and expressed as the transcript copy number per 230 000 transcript copies of GAPDH.
Histochemical and immunohistochemical analyses of the conventional mice
The fixed sample was embedded in paraffin, and 4-µm-thick serial paraffin sections were prepared. The sections were then stained with hematoxylin–eosin (H&E), and the presence of the Paneth cells was evaluated by light microscopy. The following scoring system was used: 0=no Paneth cells in the cross-section, 1=Paneth cells in at least one crypt in the cross-section, 2=Paneth cells in c. 30–60% of crypts in the cross-section 3=Paneth cells in most crypts (more than 60%) in the cross-section. In this scoring method, the number of Paneth cells per crypt was disregarded because the apparent number could vary depending on the angle at which the crypt was cut, and the H&E staining did not permit the accurate counting of Paneth cells.
Samples from mice at several ages were subjected to immunohistochemical analysis using anti-human lysozyme antibody to confirm whether the Paneth cells were properly detected by H&E staining.
Briefly, the section was deparaffinized, rehydrated and placed in 3% (v/v) H2O2–methanol for 10 min at room temperature. The slides were then immersed in blocking solution (Non-specific Staining Blocking Reagent; DakoCytomation, Kyoto, Japan) for 5 min and incubated with the anti-human lysozyme rabbit polyclonal antibody [its specificity in the FVB/N mouse intestine is described by Bry (1994); DakoCytomation] for 60 min. Antigen–antibody complexes were detected with a secondary antibody [Histofine simple stain mouse MAX PO (R); Nichirei, Tokyo, Japan] and visualized using 3,3′-diaminobenzidine (0.5 mg mL−1 in Tris-buffered saline).
Differences in gene expression between the jejunum and ileum at each age were analyzed by either Student's or Welch's t-test according to the result of the F-test. Differences between conventional and germ-free mice were analyzed by two-way anova (factors of variation were presence/absence of luminal bacteria and location in the small intestine). When significant differences were detected, Tukey–Kramer post hoc comparison was used. Differences among means were considered significant at P<0.05, and 0.05<P<0.1 was regarded indicative of a trend. All data were analyzed using statcel2 (OMS, Saitama, Japan), which is an add-in application for Microsoft Excel (Seattle, WA).
Postnatal change in gene expression of cryptdin isoforms
The expression of all six cryptdin isoform genes increased from the birth to the maturation as reported previously (Ouellette et al., 1989; Mallow et al., 1996; Darmoul et al., 1997) (Fig. 1). Almost same results were obtained by normalization to the other housekeeping gene, β-actin (data not shown).
Cryptdins 1 and 6 were highly expressed in both the jejunum and the ileum of the fetuses compared with the other isoforms. The log transcript copy numbers of cryptdins 1 and 6 in the jejunum of the fetuses were 3.60±0.32 and 3.58±0.18, respectively, and those in the ileum were 3.88±0.18 and 3.87±0.17, respectively.
For three isoforms (cryptdins 2, 4 and 5), a marked increase in gene expression was observed in both the jejunum and the ileum from prebirth until around 14 days of age; subsequently expression levels increased gradually. Gene expression of cryptdin 3 in the ileum increased markedly and rapidly after the birth. The expression level of ileal cryptdin 3 increased c. 32-fold from the fetuses to day 0. The increases in the gene expression of cryptdins 1, 3 and 6 were gradual throughout the experimental period compared with the other three isoforms. The gene expression level of all cryptdin isoforms reached the adult level (40 days old) by 4 weeks of age, as described in Ouellette (1989).
The genes that increased remarkably from prebirth to maturity were those of cryptdins 2 and 5. Expression of the cryptdin 2 gene in the jejunum and ileum increased c. 645- and 537-fold from prebirth to maturation, respectively. In the case of cryptdin 5 gene, expression in the jejunum and ileum increased c. 1700- and 4000-fold, respectively.
Interestingly, the significant differences in gene expression between the jejunum and ileum began to be detected intensively from around day 14 for most isoforms (Fig. 1). This difference between the jejunum and ileum in gene expression was observed until maturity (40 days old) for cryptdins 2 and 5, whereas it became very small after day 25 in the cases of cryptdins 1, 3 and 6.
A difference between the jejunum and ileum was already observed at the prebirth stage in the case of cryptdin 4 (P=0.057).
Gene expression of each cryptdin isoforms in conventional and germ-free mice
In all cryptdin isoforms, a significant difference in gene expression was observed between the conventional and germ-free mice (Fig. 2). Differences between the conventional and germ-free mice tended to be detected after weaning (day 22) for all isoforms except for cryptdin 2 and became significant at 40 days of ages for most isoforms (P=0.07 for cryptdin 6). Cryptdin 2 was expressed at a remarkably higher level in conventional mice than in germ-free mice at all ages evaluated, and the differences were mostly significant.
For all isoforms, significant differences between the jejunum and ileum were detected in the germ-free mice aged 19 days, and there was a tendency for such difference in most 22-day-old isoforms. However, the positional difference was small for 40-day-old mice for cryptdins 1, 3 and 6; this was similar for those isoforms in the conventional mice.
Interestingly, two-way anova revealed significant interaction between the presence and absence of luminal bacteria (conventional vs. germ-free) and the location in the small intestine (jejunum vs. ileum) for the gene expression of cryptdins 4 and 5 at 40 days of age. A similar tendency was observed for the gene expression of cryptdin 2 at 40 days of age (P=0.07).
Postnatal appearance of the Paneth cells in the conventional mice
The polyclonal antibody for human lysozyme reacted with lysozyme in the Paneth cells of BALB/c mice as well as FVB/N mice (Bry et al., 1994) (Fig. 3a). Immunohistochemical staining confirmed that our detection of Paneth cells by H&E staining was acceptable.
Both in the jejunum and in the ileum, the Paneth cells first appeared around 6 days of age (Fig. 4). The presence of the Paneth cells was confirmed visually in all individuals after 8 days (Fig. 3b and c). Thereafter, the percentage of crypts that contained the Paneth cells increased progressively, and c. 30–60% of crypts in the cross-section developed the Paneth cells at 14–18 days of age (Fig. 3d and e). In the ileum, the percentage of the crypts containing the Paneth cells further increased from 18 to 19 days of age and after this age, the Paneth cells were observed in most crypts (more than 60% of crypts in cross-section) (Figs 3g and 4). At 21 days of age, the percentage of crypts in the jejunum with the Paneth cells was similar to that in adults (Figs 3f and 4). Although the percentage of crypts with the Paneth cells reached approximately the adult level around the time of weaning, it was easily observed, especially by immunohistochemical staining, that the number of Paneth cells per crypt was still increasing after weaning as reported previously by Bry (1994) (Fig. 3).
α-Defensins (genetically encoded antimicrobial peptides) are thought to play an important role in maintaining the small intestinal environment (Ouellette & Bevins, 2001). Human Paneth cells express mRNAs for two α-defensins, HD-5 and HD-6, and mice express numerous Paneth cell α-defensins that have been termed ‘cryptdins’ for ‘crypt defensin’ (Ouellette et al., 1989).
In this study, the postnatal development of each of the six cryptdin isoforms, which have been purified as peptides, was assessed by evaluating the expression of their genes. At the same time, the expression level of each cryptdin isoform gene was compared between conventional and germ-free mice at various ages to further assess the role of luminal bacteria in the expression of each cryptdin isoform.
The results show that two patterns of postnatal development of the genes expression are present among the cryptdin isoforms. One group of cryptdins showed a gradual increase and the other showed a rapid increase in gene expressions after birth. The two patterns were generally categorized as follows: for cryptdins showing the former pattern, the transcript copy number in newborn mice was more than 100 and increased less than 500-fold between birth and maturation. For cryptdins showing the latter pattern, 100 or fewer copies were present on prebirth, and there was an increase of at least 500-fold between birth and day 40. The increase in the gene expression of cryptdins of this second group was coincident with the increase in the histochemical score for the presence of Paneth cells, especially up to a score of 2 (i.e. until c. 60% of crypts contained the Paneth cells).
In this study, cryptdins 1, 3 and 6 showed a gradual increase in gene expression and cryptdins 2 and 5 a rapid increase. Interestingly, cryptdin 4 in the ileum showed the former pattern whereas the same isoform in the jejunum showed a pattern close to the latter pattern (more than a 500-fold increase from 1 to 40 days of age, but not from the fetal stage).
Although the kind of cells expressing cryptdins before the appearance of the Paneth cells is unclear from this study, Bry (1994) has demonstrated, using an antibody that reacts with cryptdins 1, 2, 3 and 6, that goblet-like epithelial cells in the small intestine of newborn mice, which stain for UEA-1, express cryptdin peptides. The cryptdin genes detected in the early stage of life might also have been expressed in this kind of non-Paneth type cells in this study. The role of these cells in the defense of the small intestine in newborn mice needs to be further elucidated.
The difference in gene expression between the jejunum and ileum was more obvious for the isoforms that showed a rapid increase in gene expression, and the same tendency was seen in germ-free mice. On the other hand, significant differences between the two locations were also found for the isoforms that showed a gradual increase in gene expression in both conventional and germ-free mice between 14 and 24 days of age. This difference may simply reflect the difference between the jejunum and ileum in the proportion of crypts that contain Paneth cells during this period, as shown in the Paneth cell appearance scores (Fig. 4).
The expression of all cryptdin isoform genes was clearly higher in germ-free mice than in the conventional fetal and newborn mice at each of the ages evaluated. For instance, the log transcript copy number of cryptdin 4 genes in the jejunum of 13-day-old germ-free mice was 3.38±0.18 whereas that in fetuses of the conventional mice was 1.35±0.56 (Fig. 2). These facts indicate that the genes for all cryptdin isoforms developed along with progress toward maturity even in the absence of luminal bacteria. Indeed, the Paneth cells were actually observed in the germ-free mice (data not shown). These results are consistent with those of Ouellette (1989), which demonstrated the existence of the Paneth cells and the expression of cryptdin genes in adult germ-free mice.
The fact that cryptdins do not require the microbial stimulation for the increase in their expression implies the great importance of cryptdins in small intestinal defense, because we confirmed that the expression of Toll-like receptor 2, the other innate immune-related factor, was absent from the small intestine of adult germ-free mice, whereas this was not the case in conventional mice (R. Inoue, unpublished data). A broad spectrum of antimicrobial peptides may be prepared constitutively in the small intestine to enable a quick reaction to bacterial infections.
However, a significant difference between conventional and germ-free mice was detected in the expression of all cryptdin isoform genes. These differences became significant after weaning for all isoforms, especially in the ileum. Moreover, a significant two-way interaction was detected in the gene expression levels of cryptdins 4 and 5 at 40 days of age, and a similar tendency was also found in the gene expression of cryptdin 2.
These findings suggest that luminal bacteria could increase the expression of cryptdin genes and that they contribute to and/or enhance the differences between the jejunum and ileum in gene expression observed for cryptdins 4, 5 and probably cryptdin 2.
It is well known that ileum is colonized more abundantly with bacteria than the jejunum (Uhlig & Powrie, 2003). Moreover, luminal bacteria increase significantly during the weaning period, in number as well as variety (Mackie et al., 1999; Inoue & Ushida, 2003). This may explain the positional (jejunum vs. ileum) differences, and the differences between the conventional and germ-free mice became obvious around the time of weaning.
Two possibilities should be considered as the reason for the differences in the cryptdin gene expression between the conventional and germ-free mice. One is that the luminal bacteria stimulated the gene expression of cryptdins per Paneth cell, and the other is that the luminal bacteria increased the number of Paneth cells per crypt. Salzman (1998) reported that the number of Paneth cells per crypt was increased in the small intestine of necrotizing enterocolitis patients, indicating the number of Paneth cells is changeable depending on the luminal environment. Because accurate counting of the Paneth cell number was not performed in this study, both possibilities still remain. Further studies such as the study using ‘conventionalized’ animals are required to identify the precise mechanism of the increase in the gene expression of cryptdins in conventional mice.
However, the difference in the gene expression between the jejunum and ileum after the maturation should be carefully considered, especially in the case of cryptdin 4, because a significant positional difference was also observed in the germ-free mice. Furthermore, in the case of cryptdin 4, a notable positional difference was detected in the fetuses. Considering the fact that two of three isoforms (cryptdins 2 and 4) that exhibited positional differences are thought to have greater antimicrobial activity than the other isoforms, it may be suggested that the positional differences in the gene expression of cryptdins 2, 4 and 5 are ontogenetically regulated.
In conclusion, this study suggests that the cryptdin isoforms differ in their patterns of postnatal development and that the isoforms fall into two groups based on these patterns. This study also suggests that the gene expression of all cryptdin isoforms is affected by the presence of luminal bacteria. Furthermore, luminal bacteria appear to contributed to and/or enhance the difference between the jejunum and ileum in the expression of the genes for cryptdins 4, 5 and perhaps cryptdin 2. Although it seems that the Paneth cells contain a large number of copies of cryptdin genes even without the bacterial stimulation, which may be a mechanism to provide a swift response to the infections, the expression level vary according to antigenic milieu. In fact, it has been demonstrated that the gene expression of cryptdins 1, 4 and 5 is notably changed before and after the conventionalization of germ-free mice (Ogawa et al., 2000). The results of this study further imply that each cryptdin isoform plays a distinct role in the defense of the small intestine and is regulated by different factors, especially in the developing intestine.
Studies focusing on the roles of each cryptdin isoform (including any isoforms that have not yet been purified and the factors regulating them) are required to clarify the role of α-defensins in the small intestine more precisely.
The authors thank M. Harada, R. Sato, Y. Nakamoto, M. Nishikawa and M. Tanimura for their technical assistance.