INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

HIGH- AND LOW-MOLECULAR-weight kininogens (HKG and LKG), the precursors of endogenous kinins, are generated by differential splicing of a primary transcript encoded by a single gene called K-KG (18). HKG is the preferred substrate for plasma kallikrein, whereas LKG is preferentially cleaved by tissue kallikreins, a multigene family of serine proteases (for a review, see Ref. 2). The rat liver also expresses an LKG-related protein of unknown function called T-KG. Unlike LKG, however, T-KG is not cleaved by plasma or tissue kallikreins. Studies in adult kidneys indicate that LKG mRNA is expressed in the collecting ducts (14, 17), suggesting that kinin-producing cells are of ureteric bud origin. However, evidence for the production of kininogen and kinins in the developing kidney remains lacking.

It is now established that endogenous kinins modulate renal hemodynamics and urinary sodium excretion (19, 26, 27). Furthermore, transgenic mice harboring the human tissue kallikrein gene are hypotensive (30). Conversely, mice with targeted disruption of the bradykinin B₂ receptor gene manifest decreased renal blood flow and salt-sensitive hypertension (1), thus providing direct evidence for the involvement of kinins in renal function and blood pressure homeostasis.

Although the role of kinins in the adult animal is in great part understood, the role of kinins during early development is virtually unknown. Initial studies by Robillard and colleagues (25) found a significant correlation between the developmental changes in renal blood flow and renal kallikrein activity. Studies from our laboratory utilizing a selective B₂-receptor antagonist have shown that endogenous kinins, acting via B₂ receptors, counteract angiotensin II-mediated renal vasoconstriction in the developing rat (12). Furthermore, there is evidence that the activation of kinin B₂ receptor gene expression in the neonatal kidney is involved in maintaining the high-proliferative activity during the later stages of nephrogenesis (31). Other studies performed in an inbred rat strain with low renal kallikrein have shown that these rats manifest reduced renal growth and poor glomerular development (23). The presence of the mRNAs and proteins for tissue kallikrein (8, 11, 29), kininase II (angiotensin-converting enzyme) (32, 33), and bradykinin B₂ receptors (10) in the fetal and newborn kidney provides further support for the involvement of kinins in kidney maturation. However, evidence that the tissue kallikrein substrate, LKG, is expressed in the developing nephron is still lacking. Accordingly, the present study was designed to examine the ontogeny and intrarenal localization of LKG during rat nephrogenesis.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Time-dated pregnant and adult male Sprague-Dawley rats (200-250 g) were purchased from Charles River Breeding Laboratories (Wilmington, MA) and fed regular rat chow (Purina 5012; Ralston-Purina, St. Louis, MO) and tap water until the day of study.

Detection of LKG mRNA by reverse transcription-polymerase chain reaction. Total kidney RNA from fetuses (gestation days 16, 18, and 20), newborn (days 1 and 5), weanling (day 20), and adult rats was extracted by the guanidinium isothiocyanate protocol of Chomczynski and Sacchi (7). LKG mRNA was detected by RT-PCR, and liver RNA served as a positive control. The RT mixture (12 µl) contained 3 µg of total RNA, 1 ng of random hexamers, 1 µl of 10 mM dNTP, 2 µl of 10× RT buffer [200 mM Tris · HCl (pH 8.4), 500 mM KCl, 15 mM MgCl₂], and 200 U of Moloney leukemia virus reverse transcriptase Superscript II (GIBCO-BRL, Life Technologies). RT was performed using Perkin-Elmer Gene Amp PCR System 2400 (Cetus Instruments, Norwalk, CT) at 37°C for 50 min, and the RT enzyme was inactivated by heating at 70°C for 10 min. The template mRNA was degraded by incubating the samples with 2 units of Escherichia coli RNase H at 37°C for 20 min. cDNA was amplified from 25% of RT mixture by PCR, using LKG-specific primers (25 pmol of each), 1 unit of thermostable Taq polymerase, 5 µl of 10× PCR buffer, and 1 µl of 10 mM dNTP (35 cycles at 94°C for 5 min, 55°C for 1 min, 72°C for 1 min). This was followed by a 7-min elongation period at 60°C. Exponential amplification of LKG cDNA was documented at 35 cycles of PCR. The primers used were 5' ACATCACAGGTGGTTGCTGGA 3' (upstream), corresponding to nucleotides 918-938 of K-KG mRNA (exon 7), and 5' TGAGAGTCTGCCCTTGTACT 3' (downstream), corresponding to nucleotides 1215-1234 of K-KG mRNA (exon 11) (15). Because the sequence encoded by exon 11 is only present in LKG mRNA, the RT-PCR product (314) is specifically derived from LKG message. This was further confirmed by Southern blot hybridization to a radiolabeled LKG-specific oligonucleotide, 5' TCCGAAAAAGTGAGCTCCACG 3'.

As an internal standard for the efficiency of amplification, one-fifth of the RT product was also used to amplify glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA sequences. The upstream primer (5' AATGCATCCTGCACCACCAACTGC 3') corresponds to nucleotides 524-547, and the downstream primer (5' GGCGGCCATGTAGGCCATCTGGAG 3') is complementary to nucleotides 1055-1078 of rat GAPDH cDNA. PCR conditions were as follows: 94°C for 5 min, 55°C for 1 min, and 72°C for 7 min. The predicted size of the PCR product is 555 bp. Southern blots were hybridized to ³²P-labeled GAPDH cDNA. Relative band intensities were determined by scanning densitometry (Ultroscan, LKB Biotechnology).

Western blot analysis. Kidneys from fetal, neonatal, weanling, and adult rats were removed, rinsed thoroughly, and homogenized in PBS (pH 7.2) on ice. Homogenate was centrifuged at 600 g for 20 min (4°C), and the supernatant was treated with deoxycholate (0.5% wt/vol) for 30 min at room temperature and centrifuged at 20,000 g for 30 min at 4°C. The samples were subsequently assayed for protein concentration, according to the method of Lowry et al. (20), using bovine serum albumin (BSA) as a standard. Aliquots of kidney extracts (50 µg total protein) were separated by 3-12% stacking SDS-PAGE [12% acrylamide, 5 mM bis (Bio-Rad Laboratories, Richmond, CA), 0.38 M Tris · HCl, pH 8.8, 0.1% SDS, 0.1% ammonium persulfate, and 0.025% N,N,N',N'-tetramethylethylenediamine (Bio-Rad)]. Rainbow molecular weight markers (14,300-200,000; Amersham, Arlington Heights, IL) were used to determine approximate molecular weights. Electrophoresis was carried out at 60 V for 3 h in duplicate. One gel was stained with 0.1% Coomassie blue stain R250 to visualize the protein bands. The proteins from the second gel were electrophoretically transferred to nitrocellulose in a Genie Electrophoretic Blotter (Idea Scientific, Corvallis, OR) at 20 V for 90 min. Transfer buffer is Tris-glycine (25 mM Tris, 0.2 M glycine), pH 8.3, in 20% methanol. After transblotting, the nitrocellulose was washed in PBS, pH 7.2, and incubated overnight at 4°C in PBS containing 3% BSA as a blocking agent. After rinsing in 0.05% Tween 20 in PBS for 20 min at room temperature, the nitrocellulose was incubated with a polyclonal anti-rat LKG antibody (from Dr. Julie Chao) diluted 1:1,000 in PBS-Tween 20 for 2 h at room temperature. This antibody recognizes LKG but not HKG (5, 6). In other experiments, we used a monoclonal bradykinin-directed anti-KG antibody (1D₁₀), which recognizes the kinin moiety in all KGs (5). The nitrocellulose blots were then washed in PBS-Tween 20, followed by PBS alone, and incubated at room temperature for 1 h with anti-mouse IgG peroxidase conjugate (Sigma Chemical, St. Louis, MO; diluted 1:2,000 in PBS-Tween 20 with 3% BSA). After blots were washed in PBS, visualization of antigen-antibody reaction products was accomplished by enhanced chemiluminescence (ECL, Amersham) per manufacturer's instructions or by 4-chloronaphthol, as previously described (9). Relative band intensities were determined by scanning laser densitometry (Ultroscan, LKB Biotechnology).

Immunohistochemistry. Twenty-four rat fetuses (gestation ages 15 and 17-20 days), eight newborns (days 1 and 5), and four adult rats were examined for the immunolocalization of LKG. Tissues were fixed from 4-12 h in Bouin's fixative (4% paraformaldehyde, 1% picric acid, 5% glacial acetic acid), dehydrated, and embedded in paraffin blocks. Serial 5- to 7-µm sections were immunostained for KG by the immunoperoxidase technique. After deparafinization and hydration, the sections were washed in PBS for 5 min. Quenching of endogenous peroxidase activity was achieved by incubating the sections for 30 min in 0.3% H₂O₂ in methanol. After washing in PBS for 20 min, the sections were incubated with normal horse serum for 20 min. Thereafter, the sections were incubated for 90 min with LKG antibody diluted 1:2,000 in PBS containing 1% BSA. Sections were rinsed in PBS for 10 min and subsequently incubated with IgG-biotinylated antibody for 30 min. Subsequently, the sections were washed in PBS for 10 min and incubated for 45 min with the avidin-biotinylated horseradish peroxidase complex (Vectastain ABC reagent; Vector Laboratories, Burlingame, CA) and exposed for 3-5 min to 0.1% diaminobenzidine tetrahydrochloride and 0.2% H₂O₂. The sections were then washed in tap water and counterstained with hematoxylin or periodic acid-Schiff (PAS) as a marker of proximal tubular brush border. As negative controls, the primary antibody was omitted or replaced by nonimmune serum.

Immunostained sections were reviewed by light microscopy at ×100 and ×200 magnifications. The segmental localization of LKG immunoreactivity in developing and adult kidneys was recorded for each age group of rats. In addition a semiquantitative score was obtained for the intensity of staining as follow: 0, no staining; +, weak; ++, moderate; and +++, strong staining. An average of 2-3 sections per animal were examined, and all comparisons were done on tissue sections that were immunostained during the same run.

Tissue kininogen and kinin RIA. Kidney KG levels were measured by a direct RIA (6) on pooled tissue samples from 4-5 animals in each group. The kinin RIA was performed in individual kidney tissue samples, as described by Campbell et al. (3). Briefly, the rats (4 adults, 12 five day old) were killed by decapitation without prior anesthetic, and the kidneys rapidly removed and weighed. The kidneys were immediately homogenized in 4 M guanidine thiocyanate (GTC) and 1% trifluoracetic acid (TFA) in water at room temperature. The time delay between decapitation of the animal and kidney tissue homogenization was less than 60 s. To obtain sufficient kidney tissues from young rats, four pools (3 rats each) were made. Kidneys were homogenized in 20 ml GTC/TFA. Tissue homogenate was sonicated briefly and then centrifuged at 5,000 g for 20 min.

For the extraction of bradykinin from tissues, 10 ml of the supernatant from each homogenate was extracted on a Sep-Pak C₁₈ cartridge (Waters Chromatography Division, Milford, MA). The eluate was collected into a siliconized 13 × 100-mm bortosilicate glass tube and evaporated to dryness in a vacuum centrifuge. Each extract was then dissolved in 1 ml of 1 M hydrochloric acid and was extracted twice with 1 ml diethyl ether. The extracts were then evaporated to dryness again before performing the RIA (3). A rabbit polyclonal COOH-terminal-directed bradykinin (BK)-(1-9) antibody was used in the RIA. It has been shown previously that the predominant BK peptides in kidney are BK-(1-7) and BK-(1-9), with low levels of BK-(1-8) and BK-(4-9) detectable (3). Monoiodinated ¹²⁵I-tyr-BK-(1-9) was used as a tracer for the COOH-terminal-directed RIA. The standard for the COOH-terminal RIA was BK-(1-9).

Statistical analysis. Comparisons between groups were performed by one-way ANOVA or unpaired t-test. P < 0.05 was considered significant.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Developmental changes in renal LKG protein and mRNA. As shown in Fig. 1, renal LKG levels, measured by Western blotting and a specific RIA, are elevated in the fetus and newborn and decline postnatally. Peak LKG levels are observed during the perinatal period and are ~40-fold higher than the levels observed in the adult kidney. Western blot analysis also shows that the KG antibody is specific for LKG (68 kDa). Immunoblotting with a BK-directed anti-KG antibody revealed qualitatively similar results (Fig. 1B). The presence of a 68-kDa doublet is probably related to the codetection of LKG and T-KG because the utilized antibodies recognize both molecules.

View larger version (35K): [in this window] [in a new window]   Fig. 1.   Developmental changes in renal immunoreactive low-molecular-weight kininogen (LKG) assessed by Western blotting (A and B) and RIA (C). LKG protein levels are high in the developing kidney and decrease during postnatal maturation. E, embryonic day; D, postnatal day. B: 25-kDa band is probably a degradation product.

To determine whether LKG is synthesized in the developing kidney, we used a sensitive RT-PCR assay to detect LKG mRNA sequences. The RT-PCR strategy allows the distinction of LKG from HKG and T-KG. As shown in Fig. 2, LKG mRNA is expressed in the fetal kidney as early as embryonic day 16 (E16). The abundance of LKG mRNA increases toward the end of gestation and peaks perinatally only to decline thereafter. The difference in LKG mRNA levels between day 1 or 5 kidneys and adult kidneys varied between 5- and 20-fold (3 separate assays). This temporal pattern of expression parallels that of LKG protein detected by Western blotting and RIA, indicating that the renal KG gene is developmentally regulated.

View larger version (84K): [in this window] [in a new window]   Fig. 2.   Developmental changes in renal KG gene expression evaluated by RT- PCR (top) and Southern blot hybridization (bottom) specific for LKG. LKG mRNA is expressed at relatively high levels during fetal and neonatal development. GAPDH, glyceraldehyde-3-phosphate dehydrogenase; A, adult; LV, adult liver (positive control).

Developmental changes in immunoreactive renal kinin contents. Kidney immunoreactive kinins were measured utilizing a COOH-terminal kinin RIA in 5-day-old newborn and adult male rats (n = 4/group). Renal tissue kinin contents were 2.3-fold higher in the newborn than adult rats (214 ± 9 vs. 93 ± 10 fmol/g; P < 0.05, t-test) (Fig. 3). Addition of carboxypeptidase B to the assay sample, which removes the COOH-terminal arginine of BK, abolished the immunoreactivity indicating that the antibody used in this assay recognizes a true kinin moiety.

View larger version (9K): [in this window] [in a new window]   Fig. 3.   Immunoreactive kinins measured by a COOH-terminal bradykinin RIA.

Intrarenal localization of KG in the developing kidney. Sections of liver and submandibular gland (SMG) from E15 and -17 rat fetuses, respectively, were used as positive controls for the LKG immunostaining. The rat liver synthesizes LKG as early as E11 of gestation (28). LKG immunoreactivity in the fetal liver is observed in the cytoplasm of differentiating hepatocytes in a perinuclear distribution (Fig. 4A). In the SMG, LKG is observed primarily in the ductal epithelial cells and in the developing glandular structures (Fig. 4B). The loose periductal mesenchyme expresses LKG but at lower intensity.

View larger version (152K): [in this window] [in a new window]   Fig. 4.   LKG immunolocalization in fetal tissues. A: fetal liver on E15 of gestation (hematoxylin; magnification, ×170). LKG immunoreactivity is present in the perinuclear cytoplasm of immature hepatocytes. B: fetal submandibular gland on E17 of gestation (hematoxylin; magnification, ×170). LKG is expressed in the differentiating ductal and glandular epithelium and to a lesser extent in the surrounding loose mesenchyme. C: fetal kidney on E17 (hematoxylin; magnification, ×85). UB, main ureteric bud branches are LKG negative. Bulk of LKG immunostaining is seen in the nephrogenic zone. D: same fetal kidney shown in C (magnification, ×340) showing two immature nephrons expressing LKG in the differentiating terminal ureteric branches [future collecting ducts (CD); C, inset]. Scattered LKG immunoreactivity is present in the loose mesenchymal stroma.

In the developing kidney, three temporal stage-specific patterns of LKG expression were identified during nephrogenesis. During early-to-mid-fetal nephrogenesis (E15-17), LKG immunoreactivity is mainly observed in the terminal branches of the ureter (future collecting ducts) (Fig. 4, C and D). Only few LKG-positive cells are observed within the main branches of the ureter. LKG is not observed in the nephrogenic mesenchyme but is expressed at low levels in stromal interstitial cells of the cortex and medulla (Fig. 4D). PAS-counterstained kidney sections at E19 confirmed that PAS-positive maturing proximal tubules are LKG negative (Fig. 5).

View larger version (155K): [in this window] [in a new window]   Fig. 5.   Immunolocalization of LKG in the fetal rat kidney at E19 [periodic acid-Schiff (PAS); magnification, ×200]. Distinct localization of LKG in the collecting tubules. PAS-positive proximal tubules do not contain LKG immunoreactivity. Arrowheads, collecting ducts; P, proximal tubule.

During the later stages of nephrogenesis (day 5), LKG immunostaining is seen mainly in the collecting ducts (Fig. 6A). Intraglomerular LKG staining decreases in maturing glomeruli where it appears to be predominantly expressed in the capillary loops and in the peritubular capillary network (Fig. 6B). After completion of nephrogenesis, LKG assumes its classic adult-type distribution within the collecting ducts (Fig. 7B). In the adult kidney, higher concentrations of LKG antibody (1:1,000) were needed to reveal immunostaining due to the declining abundance of LKG with age. In addition to the collecting ducts, LKG immunostaining is observed in the peritubular capillary network of the outer and inner medulla (Fig. 7B). No specific staining was observed in control sections incubated with a nonimmune rabbit serum (Fig. 7A) or in the absence of the primary antibody (not shown). We also observed that the overall intensity of LKG immunostaining decreases with postnatal maturation. A summary of the segmental localization and semiquantitative assessment of the intensity of LKG immunostaining are presented in Table 1.

View larger version (126K): [in this window] [in a new window]   Fig. 6.   LKG immunolocalization in the kidney of a 5-day-old newborn rat. A: LKG is expressed primarily in the collecting ducts of deeper more mature nephrons. Some staining is seen in the peritubular capillaries. At this maturational stage, intrarenal LKG localization begins to resemble that of the adult (hematoxylin; magnification, ×200). B: mature inner cortical nephron showing a glomerulus surrounded by LKG-containing tubules. Note the basal subnuclear localization of immunoreactive LKG (arrows). LKG immunoreactivity is also seen in maturing glomeruli and peritubular capillaries (hematoxylin; magnification, ×400).

View larger version (156K): [in this window] [in a new window]   Fig. 7.   LKG immunolocalization in adult rat kidney with PAS staining. A: control section incubated with a rabbit nonimmune serum in lieu of the LKG antibody (magnification, ×200). B: LKG is localized in the collecting ducts (CD, arrowheads; magnification, ×400).

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

The present study documents that the developing rat kidney expresses LKG mRNA and protein. The ontogenetic changes in LKG expression are accompanied by significant shifts in the intrarenal segmental localization of LKG. During early nephrogenesis, LKG is mainly expressed in the terminal ureteric bud branches. As glomerulogenesis proceeds, LKG is also observed in capillary loop stage glomeruli but disappears later in fully mature glomeruli. During the later stages of nephrogenesis, LKG assumes its classic adult location in the collecting ducts.

Bradykinin is a short-lived peptide, and therefore it operates in a paracrine manner at or near its site of synthesis and release. In the adult kidney, tissue kallikrein, LKG, and kinin B₂ receptors are all expressed in the distal nephron (13, 14, 17). Emerging evidence indicates that the developing kidney may also be endowed with a local kallikrein-kinin system. The genes encoding rat tissue kallikrein and bradykinin B₂ receptors are expressed in the upper limb of S-shaped nephrons and maturing collecting ducts, respectively (8, 10). This study presents new evidence that KG gene expression is activated during kidney development, and that LKG is localized to epithelial elements destined to give rise to the collecting duct. Collectively, the results demonstrate the presence of a complete kallikrein-kinin system in the differentiating distal nephron.

The higher levels of immunoreactive kinins in developing than adult kidneys is in agreement with the observation that the intensity of LKG immunostaining is also higher in newborn than adult animals. Kininase II levels are low in the developing rat kidney and do not increase until the end of the 2nd week of postnatal life (32). In addition, there is evidence for the presence of high kininase II levels in the collecting duct, the site of kinin formation (4). Thus the elevated kinin levels in the newborn kidney are likely to result from a combination of increased production and decreased degradation.

The liver is the major source of circulating KG. HKG and LKG are encoded by differentially spliced mRNAs transcribed from a single gene called K-KG (18). The nine exons upstream of the kinin sequence code for the same amino acids in both HKG and LKG; the portion of exon 10 downstream of the kinin sequence is unique to HKG mRNA, whereas exon 11 is expressed in only LKG mRNA. LKG (68 kDa) is the substrate for tissue kallikrein, whereas HKG (100-120 kDa, depending on the species) serves as the substrate for plasma kallikrein. The rat liver expresses an additional unique KG molecule, called T-KG, which displays 90% sequence homology with LKG (2). T-KG is a product of an acute-phase gene that is stimulated by interleukin-6 in inflammatory conditions. The mRNAs encoding LKG and T-KG are differentially regulated. Both transcripts of the K-KG gene (i.e., HKG and LKG) are expressed constitutively during rat liver development (9). On the other hand, T-KG expression is induced immediately after birth (9). The present study shows that the K-KG gene is induced in the developing kidney. Thus the ontogeny of K-KG gene expression is controlled by tissue-specific factors. The molecular basis for the differential tissue-specific and developmental expression of KG is an interesting topic for future investigation. This study focused on LKG, rather than HKG or T-KG, because LKG is the main physiological substrate for renal tissue kallikrein.

The recent availability of highly potent and selective bradykinin receptor antagonists has permitted the evaluation of the role of the kallikrein-kinin system in renal and cardiovascular homeostasis. In the adult, acute kinin B₂ receptor blockade decreases renal blood flow and reduces papillary blood flow and the natriuretic response to volume loading (26, 27). The role of kinins in blood pressure regulation is further illustrated by the findings that chronic administration of angiotensin II to kininogen-deficient rats or kinin receptor antagonist-treated rats results in hypertension (22, 24) and that transgenic mice overexpressing human tissue kallikrein are hypotensive (30). In contrast, that targeted disruption of the B₂ receptor gene confers salt sensitivity (1).

The functional relevance of intrarenal kinins in the developing animal is not entirely clear. Recent studies using selective B₂ receptor antagonists have demonstrated that endogenous kinins tonically oppose angiotensin-mediated renal vasoconstriction in newborn rats (12). Also, kinin B₂ receptor blockade in neonatal (but not adult) rats suppresses renal growth and DNA synthesis (31). A role for the renal kallikrein-kinin system during fetal life has also been suggested. Gestational blockade of B₂ receptors in salt-treated pregnant rats compromises fetal metanephrogenesis and leads to adult hypertension (21). Ongoing studies in mice with genetic ablation of B₂ receptors should be informative with regard to the developmental role of kinins.

There is evidence that the intact LKG molecule may have additional functions independently from kinins. The KG molecules are potent inhibitors of lysosomal cysteine proteases, such as cathepsin B, H, and L (2). During the period of rapid cell growth and proliferation, these enzymes may be released from apoptotic cells and, if activated in an acid microenvironment, can induce substantial injury or extensive remodeling of the extracellular matrix in the immature kidney. We therefore postulate that the developmentally programmed activation of KG gene expression may represent an intrinsic defensive mechanism to protect the differentiating renal epithelial cells against endogenous cysteine proteases released from neighboring apoptotic mesenchymal cells. In support of this hypothesis, preliminary studies by Vattimo et al. (28) have suggested a role for cysteine proteinase activity in growth factor-induced tubulogenesis. Clearly, future studies are needed to test this hypothesis.

In summary, this study demonstrates that LKG is highly expressed during rat nephrogenesis. The differentiating collecting duct, the site of kinin B₂ receptor expression, is also the principal LKG-expressing segment in the developing nephron. Based on the infrequent expression of KG in the main ureteric branches or metanephric mesenchyme, we speculate that kinins are not involved in the regulation of branching morphogenesis or early mesenchymal-epithelial differentiation. On the other hand, the spatially restricted expression of LKG (along with B₂ receptors, Ref. 10) in the differentiating collecting duct system supports the idea that kinins and/or KGs may be important for distal nephron growth or the acquisition of a differentiated phenotype.