INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

THE NEONATE IS CHARACTERIZED by an immaturity in maintaining acid-base homeostasis. During the first weeks of postnatal life, increasing filtered loads stimulate the development of HCO reabsorption capacity of the proximal tubule (9, 26, 29). Several processes involved in urine acidification, such as the expression of proton pumps (3, 15), Na⁺/H⁺ exchangers (2, 4, 27), and chloride-base exchange (28) undergo maturational changes. There are also maturational changes in the distal nephron during development (16, 21-22, 29), which contribute to the overall maturation of urine acidification and electrolyte homeostasis. In addition to maturational changes in structure and transporters, there are increases in the nephron activities of alkaline phosphatase, glucose-6-phosphatase, and 5'-nucleotidase (32). It is likely that there are maturational increases in several of the carbonic anhydrase (CA) enzymes, including CA II (5, 7) and CA IV (25); however, the changes in specific nephron segments have only been examined for CA II (12). Immaturity of CA enzymes could impair the infant's ability to excrete an acid load and contribute to some forms of metabolic acidosis in prematurity. Data concerning the maturation of CA IV expression in specific nephron segments, particularly in the medulla, are presently not available.

CA, a zinc metalloenzyme, catalyzes the hydration of CO₂ and the dehydration of carbonic acid. Of the 14 isoenzymes isolated thus far, two of the major renal isozymes are cytosolic CA II, which accounts for 95% of activity and membrane bound CA IV, which accounts for much of the membrane-associated CA activity. CA IV facilitates HCO reabsorption in the proximal tubule by catalyzing the dehydration of intraluminal carbonic acid, which is formed by the combination of filtered HCO with secreted protons. By immunohistochemistry in the mature kidney, CA IV expression is found primarily in the proximal tubule and inner medullary collecting ducts, as well as in the apical membranes of type A (H⁺-secreting) intercalated cells (24). In addition, apical membrane CA activity has been identified in proximal convoluted tubules (14), medullary collecting duct (13, 20), as well as along the inner stripe of the rabbit outer medullary collecting duct (31) and in the inner medullary collecting duct (33). CA IV mRNA has also been demonstrated in these segments by RT-PCR (30).

There have been few studies addressing the maturational changes of CA isoenzymes in the developing kidney (5, 12, 25). A previous maturational study (25) of CA IV protein expression (by immunoblot) in rabbit kidney used an antibody that was not satisfactory for immunohistochemistry. The purpose of the present study was to examine the development of CA IV mRNA and protein expression in specific nephron segments of maturing rabbit kidneys by comparing results from Northern blots, in situ hybridization, and immunohistochemistry. These studies were accomplished by using a newly developed goat anti-rabbit CA IV antibody (24); we also used a full-length cDNA coding for rabbit CA IV (34), which allowed us to perform Northern blots on maturing cortex and medulla and to generate cRNA probes for in situ hybridization.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals. Pregnant New Zealand white rabbits, purchased from Hazelton-Dutchland farms (Denver, PA) were allowed to deliver in our animal quarters to provide newborns (1-7 days of age). Litters of 1- to 2-wk-old pups were purchased with their mothers and allowed to grow to maturity in our facility. The diet for the adult animals (weight: 1.8-3.1 kg) consisted of standard laboratory chow (Purina Mills, Richmond, IN) and freely accessible tap water. The pups were fed by and raised with their mothers.

Preparation of total RNA from kidney and other tissues. As previously described (34), coronal slices of 1- to 2-mm thickness were dissected into cortical and inner medullary zones and snap frozen in liquid nitrogen. In the first 3 wk of life, outer medulla cannot be readily distinguished from inner medulla (25, 32), so that the medulla was cut 1 mm below the cortex. Approximately 1 mm of the papillary tip was removed from animals of all ages.

Northern blot hybridization. We performed Northern analysis as previously described (34). Briefly, 10 µg of total RNA from cortex and medulla were denatured in 3% formaldehyde, size fractionated on 1.5% agarose gels, transferred to nylon filters (Zeta-Probe, Bio-Rad), and hybridized with DNA probes radiolabeled with ³²P[dCTP] by random hexanucleotide extension (Amersham, Arlington Heights, IL). The DNA probes were rabbit CA IV (34) and rat -actin (housekeeping gene) (17), which were labeled to a specific activity of 1.5 × 10⁹ counts · min¹ · µg¹. Autoradiography was performed using Kodak XAR film at 80°C for 1-3 days.

In situ hybridization. The perfused kidneys were fixed in 4% paraformaldehyde overnight at 4°C, dehydrated through a graded ethanol series, followed by incubation in xylene. After impregnation in paraffin wax, the tissue was cast into blocks. Serial sections of tissue of ~5- to 8-µm thickness were cut on a microtome, placed on positively charged slides (Superfrost Plus, VWR Scientific, Piscataway, NJ), and stored at 80°C.

1

1

1

1

Immunohistochemistry. Whole kidneys were perfused and harvested as described above. The kidneys were cut into three sagittal sections and fixed overnight in 2% paraformaldehyde, 75 mM lysine, 10 mM sodium periodate (PLP) at pH 7.4 and then paraffin embedded. Sections (4-6 µm) were deparaffinized with Propor (Anotech, Battle Creek, MI) and hydrated in a decreasing ethanol series. Hydrogen peroxide 0.3% was used to remove endogenous peroxidase. The membranes were permeabilized using 0.3% Triton X-100 and tissue blocked with 10% horse serum. The primary antibody was an affinity-purified goat anti-rabbit CA IV peptide antibody (24) in a concentration of 1:125 in 5% horse serum. The antibody had been generated against the NH₂-terminal amino acids numbered 73-88 (YDQREARLVENNGHSV) of the deduced 308-amino acid sequence of rabbit CA IV. The secondary antibody was biotin horse anti-goat (Vector, Burlingame, CA) in a concentration of 1:450 in 5% horse serum. According to manufacturer's instructions, avidin-biotinylated horseradish peroxidase (Vectastain Elite ABC kit) was applied followed by reaction for 10 min with the substrate diaminobenzidine tetrahydrochloride for brown color development. In some cases, no counterstain was performed to better resolve the faint staining in the immature kidneys. Competition studies were performed by preincubating the antibody overnight at 4°C with the immunizing peptide at 10-fold molar excess.

Analysis and statistics. The intensity of signals in 24- to 72-h autoradiograms was analyzed by scanning densitometry (AlphaImager 2000, v.3.3, Alpha Innotech, San Leandro, CA, or SigmaGel, Jandel, San Rafael, CA). To compensate for differences in quantity of total RNA on each lane of the membrane, each CA IV value was normalized to its respective value of -actin. To allow for differences in the intensity of signals among various Northern analyses, one adult kidney cortex was run with each maturational study. The adult cortex was set as 100%, and all other samples were expressed as a percentage of this reference.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Northern analysis of CA IV mRNA expression. The maturational expression of CA IV in zones of the kidney was examined by Northern analysis. As shown in the representative blot depicted in Fig. 1, a single band of ~1.6 kb was detected at all ages in renal cortex and medulla. Densitometric analysis revealed an increase of steady-state mRNA expression after the first 2 wk of life in cortical tissue (Fig. 2). The expression in the first 2 wk showed significant variability among individual pups and no tendency to increase with age. After 2 wk of age there was a maturational increase; the levels at 5 wk of age were not significantly different from adult values.

View larger version (35K): [in this window] [in a new window]   Fig. 1.   Northern analysis of 10 µg of total RNA from maturing kidney cortex (Ctx) and inner medulla (IM). Blots were probed for rabbit carbonic anhydrase (CA) IV mRNA (top) and then for actin (bottom) to control for equal loading. The left blots show that adult cortex expresses more CA IV than does the medulla and more than either cortex or medulla of the 4-day-old rabbit, when normalized to actin. The right-hand blots show that medullary CA IV expression in 18- to 27-day-old rabbit kidneys appears more abundant than in adult medulla and is consistently greater than the expression of the cortex at the same age.

View larger version (10K): [in this window] [in a new window]   Fig. 2.   Densitometric analysis of total RNA from maturing rabbit kidney cortex. Densitometry was performed on the CA IV and actin signals, and ratios were determined for each tissue for each blot. These ratios were normalized to the adult (set as 100%). The asterisks indicate statistically different mean ratios from adult values by Scheffé's and Tukey-Kramer tests, P < 0.05.

View larger version (11K): [in this window] [in a new window]   Fig. 3.   Densitometric analysis of total RNA from maturing rabbit kidney medulla. Densitometry was performed on the CA IV and actin signals, and ratios determined for each tissue for each blot. These ratios were normalized to the adult (set as 100%) that was run in the same study. The asterisks indicate statistically different mean ratios from the 5-wk medullary value by Scheffé's and Tukey-Kramer tests. The adult value was significantly different from the 5-wk value only by Scheffé's test, P < 0.05.

In situ hybridization. To localize CA IV mRNA expression patterns in the immature kidney, we performed in situ hybridization. Figure 4 depicts the expression pattern in representative in situ hybridization slides at three different maturational ages. At 5 days of age, there was little CA IV mRNA expression in the outer cortex (top left) and medulla (middle left), with the glomeruli, inner medulla, and papilla (bottom left) being completely negative for signal (left panels). Some detectable signal was noted in the deep cortex at the cortico-medullary junction (arrowheads). By 21 days (middle) the expression at the cortico-medullary junction was heavier (top middle, arrowhead), but the medullary rays did not express much CA IV mRNA (arrow). Definite expression was detected in the inner medulla (middle), most likely over medullary collecting ducts, with negative expression in the terminal inner medulla and papilla (bottom middle). Compared with the positive findings with the anti-sense CA IV probe in the inner medulla at 3 wk of age, the sense probe gave no signal (Fig. 5, top panels). In adult kidneys (Fig. 4, right panels), cortical expression was primarily along the medullary rays (top right panel) and at the corticomedullary junction. Medullary expression (Fig. 4, middle right panel) was most abundant in the initial inner medulla, and primarily in the collecting ducts, and less heavy in the outer medulla (top right panel below medullary rays); the terminal inner medulla and papillary tip remained negative (bottom right panel).

View larger version (148K): [in this window] [in a new window]   Fig. 4.   Low-power views of in situ hybridization studies of CA IV mRNA expression in maturing rabbit kidneys. Cortical, medullary, and papillary regions of 5-day, 21-day, and adult rabbit kidneys are visualized by using darkfield illumination. There is a faint signal for CA IV in the 5-day kidney only at the corticomedullary junction (arrowheads); expression is low in the medulla and not visible in the papillary tip. In the cortex of the 21-day kidney, the heaviest signal is in the juxtamedullary area (arrowhead) with some faint labeling along the developing medullary rays (arrow). The outer medulla has no signal (not shown), but there is strong signal over the inner medulla, except for the papillary tip. The adult shows a prominent signal over the medullary rays, minimal signal over the outer medulla, and heavy signal in the inner medulla except for the papillary tip. Bar = 250 µm.

View larger version (130K): [in this window] [in a new window]   Fig. 5.   Low-power comparisons of positive and negative signals in maturing rabbit kidneys. Top row shows in situ hybridization of 18-day kidney medulla with anti-sense (left) and sense (right) CA IV probes. The positive signal over medullary collecting ducts is not seen using sense probe. Bar in right panel = 250 µm. The second row of panels shows CA IV immunohistochemistry of 10-day kidney cortex, and the third row shows the medulla using antibody (left) and antibody plus 10-fold molar excess of competing peptide (right). The peptide completely prevented the staining pattern in the immature kidney. The fourth row of panels shows CA IV immunohistochemistry of adult kidney cortex using antibody (left) and antibody plus competing peptide (right). The peptide completely inhibited the development of the brown positive signal over the medullary rays. Bar in right panel of row 4 = 250 µm and also applies to rows 2 and 3.

Immunohistochemistry. CA IV protein was demonstrated by immunohistochemistry (brown staining in Fig. 6). The pattern of staining appeared grossly similar to that seen by in situ hybridization (cf. Fig. 4). The 3-day animal (Fig. 6, left panels) had minimal expression in the medullary rays and in the nephrogenic zone (Fig. 6A). There was definite but faint expression at the corticomedullary junction, which became more apparent by 10 days of age (Fig. 5, second row, left panel). A higher power view of the deep cortex of the 3-day-old kidney (Fig. 6C) showed faint staining in some proximal convoluted tubules and no label in the glomeruli. There was modest medullary expression (Fig. 6B, left panel, and Fig. 5, third row, left panel) but no papillary expression. This staining could be eliminated by preincubating the antibody with excess immunizing peptide (Fig. 5, second and third rows, right panels). By 18 days (Fig. 6, middle panels), there was stronger cortical expression (A), especially at the corticomedullary junction but less along the medullary rays. There was prominent expression in the medulla (B), primarily along the medullary collecting ducts (C), as in the mature kidney. In the adult animal (Fig. 6, right panels), expression was prominent along medullary rays in the cortex (A), primarily in proximal straight tubules and less so in the proximal convoluted tubules of the cortical labyrinth. Juxtamedullary proximal convoluted tubules also showed heavy staining. This staining could be prevented by preincubating antibody with peptide (Fig. 5, fourth row). There was strong expression in the inner medulla (Fig. 6B, right panel) primarily along the medullary collecting ducts (Fig. 6C). Glomeruli and papillary tip (Fig. 6D) were negative for CA IV immunohistochemistry throughout all ages.

View larger version (161K): [in this window] [in a new window]   Fig. 6.   Low-power views of CA IV immunohistochemistry in maturing rabbit kidneys. Cortical, medullary, and papillary regions of 3-day, 18-day, and adult rabbit kidneys are visualized by using brightfield illumination of the brown DAB reaction product. A: the nephrogenic outer cortical zone of the 3-day rabbit kidney is negative for CA IV, but there is some staining at the corticomedullary junction. The medullary rays are virtually negative at this age. The 18-day kidney is more heavily labeled particularly at the corticomedullary junction, and there is some staining in the medullary rays. The adult shows prominent staining of medullary rays, and signal is heaviest at the corticomedullary junction. B: the inner medulla and papilla of the 3-day old show light staining. In the 18-day old, the inner medulla is well stained particularly over the collecting ducts (bottom part of figure); the outer medulla is negative at this age (top). In the adult, the outer medulla (top) is lightly stained as it makes a transition to the heavier stained initial inner medulla. C: a higher power view of the deep cortex of the 3-day-old kidney shows faint staining of proximal convoluted tubules. Bar = 100 µm for this panel only. A low-power view of the inner medulla shows the positive-staining collecting ducts of the 18-day-old and the adult. D: the papilla was negative in the adult, as well as in the younger animals (not shown). Bar = 250 µm and applies to all other panels in this figure other than C, left.

View larger version (161K): [in this window] [in a new window]   Fig. 7.   High power immunohistochemistry of CA IV in 3-day, 18-day, and adult rabbit kidneys. A: there is no staining of the nephrogenic zone of the 3-day rabbit kidney. The 18-day old shows staining of the juxtamedullary proximal convoluted tubules with more pronounced apical staining (arrow). In the inner cortex, adult kidneys show heavier and more diffuse staining of the proximal convoluted tubules with pronounced apical staining (arrow), but also with cytosolic staining. Glomeruli are consistently negative at all ages. B: deep cortical views show some faint staining of proximal tubules in the 3-day-old (arrows) and progressively heavier labeling in the medullary rays of the 18-day-old (arrows) and adult; there is probably some basolateral staining of adult proximal straight tubules. A collecting duct type A-intercalated cell expressed apical CA IV (arrow). C: labeling of inner medullary collecting ducts increases with age; the staining in the younger animals is primarily apical (arrows), whereas in the adult the staining is more diffuse, including apical and lateral membranes and cytosol. Bar = 50 µm.

View larger version (86K): [in this window] [in a new window]   Fig. 8.   High-power views of CA IV immunocytochemistry in cortical collecting ducts from 10-day, 18-day, and adult kidneys. Occasional cells show apical brown label of CA IV (arrows) consistent with their being H+-secreting A-intercalated cells (24). The staining pattern changes from a fine apical line at 10 days (top, arrow) to a thicker apical label at 18 days (middle, arrows), which is more intense at maturity (bottom, arrows). Bar = 50 µm.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

CA IV is an important enzyme in renal acid-base transport. Because there are large maturational increases in proximal tubule (26) and medullary collecting duct (16) HCO reabsorption, it would be expected that there would be concomitant increases in cortical and medullary CA IV. Indeed, this has been recently demonstrated by immunoblotting of kidney cortical and medullary membranes from maturing rabbits (25); however, this antibody failed to show medullary expression of CA IV protein by immunohistochemistry, despite the previous demonstration of CA IV mRNA in the inner medullary collecting duct (34). In addition, there have been no nephron-specific studies of the maturation of CA IV gene expression. For these reasons, we used a newly developed goat anti-rabbit CA IV peptide antibody to perform immunohistochemistry and a CA IV cRNA probe to perform in situ hybridization on kidney sections from maturing rabbits. We also obtained densitometric data from Northern blots of cortical and medullary kidney mRNA from maturing rabbits. In general, there was excellent agreement between the in situ hybridization and immunocytochemical patterns, and these paralleled the densitometric data obtained from Northern blots of total RNA from cortex and medulla: expression of CA IV was quite modest early in postnatal life and increased after the second postnatal week, to reach mature levels early in the second month of life.

CA IV mRNA and protein were expressed by the mature proximal tubule, and the pattern showed heavy labeling of the medullary rays and juxtamedullary tubules, with less staining of the early superficial proximal tubules and even less of the S3 proximal straight tubules in the medulla. The medullary ray pattern of CA IV expression early in life was not observed, nor was there expression in the nephrogenic zone. Expression of CA IV was most evident in convoluted tubules near the juxtamedullary glomeruli, and more in the labyrinths than in the medullary rays. This immature pattern was still evident at 18-21 days of age, and the medullary rays were just beginning to show CA IV expression in the proximal straight tubules therein. This increase in cortical CA IV expression was similar to that obtained from the densitometric analysis of CA IV Northern blots of maturing kidney cortex. This is not unexpected in view of the fact that proximal tubules comprise the major cell type expressing CA IV in the kidney cortex (6, 8).

We previously showed that there was a maturational surge in juxtamedullary proximal tubular HCO absorption between 4 and 6 wk of age, such that mature levels of transport are reached by the end of this period (26). We also concluded from the morphological appearance of the maturing proximal tubules (10) that the superficial proximal tubules lagged behind the juxtamedullary tubules by at least 1 postnatal wk. Thus the gradual increase in cortical CA IV mRNA expression is probably adequate to mediate the maturational increases in transport that occur after 3 postnatal wk of age. The pattern of this maturational increase in CA IV mRNA expression agrees well with the previously reported increase in CA IV protein by Western blot (using a different antibody) (25): expression was ~20% of the adult during the first 2 postnatal wk before steadily increasing to adult levels early in the second month of life.

CA IV mRNA and protein were expressed by the mature medullary collecting duct, especially the initial inner medullary collecting duct, as has been shown previously (24, 30, 34). The neonatal medulla showed weak labeling of inner medullary collecting ducts, but this appeared nearly mature by 18 days of age. Indeed, the densitometric expression of CA IV mRNA at 5 wk of age appeared numerically higher than that observed in the adult, and the intensity of the in situ hybridization signal over the medullary collecting ducts at 18 days of age appeared to be comparable to that seen in mature kidney inner medulla. Presumably, the high level of medullary CA IV mRNA during the third to fifth week of life results in early maturation of CA IV protein expression, as shown previously (25). This pattern also reflects the early maturity of medullary collecting duct HCO absorption (16) and cell pH (23). In addition, urinary concentrating ability, another major function of the medulla, matures after 21 days of age (11), parallel to the expression of CA IV.

The membrane distribution of CA IV expression in the proximal tubule has previously been shown to be apical and basolateral (8, 24, 25). Because CA IV is a GPI-linked protein (18), it is likely to be expressed only apically (19). Perhaps, a non-GPI-linked form of CA IV or another isoenzyme of CA is expressed basolaterally and was detected using our anti-peptide antibody. Further studies are needed to determine the identity of the basolateral expression.

In the present study, the staining pattern of CA IV in the immature kidney differed from that of the adult. Whereas the pattern of CA IV labeling in the immature kidney was typically apical in proximal tubules, A-intercalated cells and in medullary collecting duct cells, in mature kidneys the staining in the proximal tubule showed apical as well as basolateral signals, and that in the medullary collecting duct cells was more apical and cytosolic, as described previously (24). Even though sections from all age groups were batched to minimize intra-assay variability, it is not likely that the change in staining pattern is due to overstaining of the mature kidney or greater drift of the reaction product. The pattern in the mature kidney did not change much after shorter color generation time (not shown). At a minimum, we can conclude that early expression of CA IV in renal epithelial cells was primarily apical, as expected for a GPI-anchored protein. Whether the appearance of basolateral and cytosolic CA IV signal in the mature kidney reflects a high rate of CA IV synthesis, a different isoform of CA IV, another membrane-bound CA, or an artifact of fixation will require further investigation.

In summary, these are the first studies to examine the maturational expression of CA IV mRNA and protein in kidney sections, and in nephron segments. There was good agreement between the results from in situ hybridization and immunohistochemistry, and these correlated well with the Northern blots. Expression of CA IV mRNA and protein was ~20% of the adult during the first 2 wk of postnatal life, before increasing in parallel with the previously observed increases in transport. The maturation of the medulla appeared to precede that of the cortex, as has been reported for a variety of functional studies.