Analysis of Kif5b Expression during Mouse Kidney Development

Recent studies showed that kidney-specific inactivation of Kif3a produces kidney cysts and renal failure, suggesting that kinesin-mediated intracellular transportation is important for the establishement and maintenance of renal epithelial cell polarity and normal nephron functions. Kif5b, one of the most conserved kinesin heavy chain, is the mouse homologue of the human ubiquitous Kinesin Heavy Chain (uKHC). In order to elucidate the role of Kif5b in kidney development and function, it is essential to establish its expression profile within the organ. Therefore, in this study, we examined the expression pattern of Kif5b in mouse kidney. Kidneys from embryonic (E) 12.5-, 16.5-dpc (days post coitus) mouse fetuses, from postnatal (P) day 0, 10, 20 pups and from adult mice were collected. The distribution of Kif5b was analyzed by immunostaining. The possible involvement of Kif5b in kidney development was investigated in conditional mutant mice by using a Cre-LoxP strategy. This study showed that the distribution of Kif5b displayed spatiotemporal changes during postnatal kidney development. In kidneys of new born mice, Kif5b was strongly expressed in all developing tubules and in the ureteric bud, but not in the glomerulus or in other early-developing structures, such as the cap mesenchyme, the comma-shaped body, and the S-shaped body. In kidneys of postnatal day 20 or of older mice, however, Kif5b was localized selectively in the basolateral domain of epithelial cells of the thick ascending loop of Henle, as well as of the distal convoluted tubule, with little expression being observed in the proximal tubule or in the collecting duct. Conditional knock-down of Kif5b in mouse kidney did not result in detectable morphological defects, but it did lead to a decrease in cell proliferation rate and also to a mislocalization of Na+/K+/-ATPase, indicating that although Kif5b is non-essential for kidney morphogenesis, it is important for nephron maturation.

Introduction chain (Kif5b) to facilitate the transport of N-cadherin-catenin complexes to adhesion junctions [16], indicating a possible involvement of Kif5b in E-cadherin transportation in epithelial cells. Finally, in colonic epithelial cells, Kif5b is localized at intact and internalized apical junctions and can mediate the disassembly/internalization of adhesion junctions (E-cadherin) and tight junctions (occludin) upon Ca 2+ depletion [22]. Identification of these cargoes in a range of epithelial cells therefore suggests that Kif5b may play an important role in renal tubular epithelial cell polarity and function.
Against this background, and as a first step in obtaining clues to the functions of Kif5b in kidney development and function in vivo, we have in this study examined the spatiotemporal expression patterns of Kif5b in mouse kidney.

Materials and Methods Animals
C57BL/6N, ROSA26 reporter line, Kif5b +/-, Kif5b fl/fl and Pax2-Cre mice were used in this study. Generation of Kif5b +/-:Pax2-Cre mice has been described elsewhere [23]. The day on which the cervical mucus plug was observed was designated as embryonic day 0.5 (E0.5) and the day on which birth occurred was designated as postnatal day 0 (P0). For each time point, three mice were obtained from separate litters. Mouse experimentation was carried out in strict accordance with the recommendations in the guide for the Care and Use of Laboratory Animals of the United States National Institutes of Health. The protocol was approved by the Committee on the Use of Live Animals in Teaching and Research (CULATR) at The University of Hong Kong (Permit Number: 2540-11). All surgery was performed under sodium pentobarbital anesthesia, and all efforts were made to minimize animal suffering. In total, more than ten litters of mutant and control mice or embryos were examined in this study.
For immunohistochemical analysis, rehydrated paraffin sections were briefly rinsed with PBS and blocked with Peroxidase-Blocking Reagent (Dako) for 10 min, and then incubated with blocking solution (5% [w/v] goat serum, 3% [w/v] BSA [bovine serum albumin] and 0.5% [v/v] Triton X-100 in PBS) for 1 hour at room temperature, followed by incubation with rabbit anti-Kif5b antibody (1:200) at 4°C overnight. After being washed with TBST (Tris-buffered saline with Tween-20), the sections were incubated with HRP-conjugated anti-rabbit secondary antibody at room temperature for 1 hour. Staining for Kif5b was undertaken using liquid DAB (3,3'-diaminobenzidine) and the substrate-chromogen solution from the EnVision+ System-HRP (DAB) kit (Dako), with hematoxylin counter-staining. Specimens on slides were observed under a light microscope (Carl Zeiss) and photographed using a Microphoto-FX microscope (Nikon) attached to a digital camera (Nikon).
Total RNA extraction and quantitative PCR Total RNA was isolated from tissues using Trizol reagent (Invitrogen). First-strand cDNA was synthesized using Superscript III reverse transcriptase (Invitrogen). Quantification of mRNA levels was performed using SYBR Premix Ex Taq (Takara). The reference gene chosen was 18S ribosomal RNA. The sequences of the primers are listed in S1 Table. Quantitation of cell proliferation in kidney Cell proliferation in kidneys was determined by 5-bromo-2'-deoxyuridine (BrdU) incorporation using a BrdU staining kit (ZYMED Laboratory). Pregnant female mice received an intraperitoneal injection of BrdU (100 μg/g body weight) 4 hours prior to being killed. Identification of BrdU-positive cells was performed by immunostaining as described previously [24].

Statistical analysis
Sigma Stat (Systat Software) was used to analyze the data which were expressed as the mean ± S.D. Comparisons between two mean values were performed by independent samples t-test. For multiple comparisons among different groups of data, significant difference were determined by one-way analysis of variance (ANOVA) followed by Tukey's test. P<0.05 was considered to be statistically significant.

Results and Discussion
Kif5b is widely expressed in tubular cells in the neonatal mouse kidney Although Kif5b is widely expressed in mouse tissues [8], it exhibits various expression levels in different cell types within a specific organ. In the nervous system, Kif5b was found to be highly expressed in olfactory sensory neurons [8], whereas in the pancreas Kif5b was predominantly localized in islets, with little expression in exocrine acinar cells in adult mice [24]. It is currently poorly documented whether the distribution of Kif5b varies among different cell types in the kidney.
To characterize Kif5b expression in the kidney, we analyzed the distribution of Kif5b in neonatal kidneys. In the newborn mouse, the kidney can be divided into three parts: the outermost part is the nephrogenic zone where nephrogenesis occurs, and the inner parts are the cortex and the medulla, respectively ( Fig 1A). Kidney cortex is composed of glomeruli and renal tubules, while the medulla comprises only tubules. The interstitium is filled among the parenchyma in the whole kidney.
Based on immunohistochemical analysis, Kif5b was found to be localized mainly in the renal tubules of the cortex and medulla, with little expression in the nephrogenic zone or glomeruli of the cortex (Fig 1A). The positive staining was absent after pre-incubating the anti-Kif5b antibody with the 18-aa synthesized peptides (S1C Fig). In the nephrogenic zone, Kif5b was expressed prominently in the ureteric bud, but it exhibited a very low level of expression in the cap mesenchyme, the comma-shaped bodies, the S-shaped bodies and the advanced S-shaped bodies ( Fig  1B). This indicates that Kif5b is expressed more strongly in structures that are of Wolffian duct origin than in MM-derived early nephron structures. We next examined the expression of Kif5b in the MM-derived late nephron structures: the glomeruli and renal tubules in the cortex and medulla. We found that Kif5b was widely expressed in the renal tubules of both cortex and medulla, such as the proximal convoluted tubules (PCT), the thick ascending limbs of the loops of Henle (TAL), the distal convoluted tubules (DCT) and the collecting ducts (CD), but the expression in glomeruli was quite rare (Fig 1C and 1D). Expression of Kif5b in the interstitial cells was also hardly detectable ( Fig 1B-1D).

The expression pattern of Kif5b changes with kidney development
To characterize the spatial dynamics of Kif5b expression during kidney development, we examined the expression pattern of Kif5b in E12.5, E16.5, P0, P10, P20 and adult kidneys. Expression of Kif5b was not detectable in the kidneys of E12.5 embryos (Fig 2A and 2B); however, in E16.5 kidneys, Kif5b was highly expressed in the ureteric bud and tubular structures (Fig 2C  and 2D). The quantitative levels of Kif5b mRNA during embryonic kidney development were determined by real-time reverse transcription PCR (qRT-PCR) ( Fig 2E). The Kif5b mRNA was not detected in the early embryonic kidney (E12.5), but its level increased significantly in the kidneys of E16.5 embryos and thereafter. Although Kif5a and Kif5b were reported to be neuronal specific [8], both of them could be detected in E12.5 kidneys. The expression levels of Kif5a and Kif5c declined significantly during later kidney development, and could not be detected at all in the kidneys of newborn mice (Fig 2E and S1D Fig) In both the cortex and the medulla, the spatial expression of Kif5b changed with development in the postnatal kidney in two aspects. First, Kif5b expression gradually became restricted to DCT and TAL (Fig 3A-3I). This alteration became obvious at P20 (Fig 3E, 3F and 3I), at which stage the renal tubules probably became functionally mature. Secondly, the subcellular localization of Kif5b changed during development. In the newborn mouse, Kif5b was evenly distributed in the cytoplasm of renal epithelial cells (Fig 3A, 3B and 3J  and 3K). Although Kif5b was preferentially expressed in some tubules during postnatal kidney development, its total expression level in the whole kidney remained unchanged during this process (S1E Fig). The presence of Kif5b in the tubules of mouse kidney indicates that Kif5b may be involved in a re-absorption and/or secretion process specific to renal tubules. Furthermore, dynamic spatial changes either intracellularly or along the length of differentiating renal tubules (Fig 3) indicate a requirement of Kif5b during kidney postnatal development. To further confirm the nephron segments in which Kif5b was expressed in adult mice, kidney paraffin sections were double labelled for Kif5b and for markers of specific nephron segments (Fig 4). Lotus tetragonolobus agglutinin (LTL) is a lectin that specifically stains proximal tubules (green) [6]. Double immunostaining showed that Kif5b was not expressed in the proximal tubule (PT). Aquaporin-2 (AQP2) is localized mainly at the apical cell membranes of the principal cells of the CD of the kidney [25]. Lack of Kif5b expression in AQP2-positive tubules revealed that Kif5b was not expressed in the CD. Tamm-Horsfall protein, a GPI-anchored glycoprotein (THP), is found to be produced by the TAL in mammalian kidney [6]. Here, positive staining of Kif5b and THP in the same renal tubules suggested that Kif5b was expressed in the TAL. Finally, the Na-Cl cotransporter (NCC) is specifically localized at the apical surface of the DCT. Since antibodies against NCC and Kif5b were both raised in rabbit, we stained these two proteins separately in two consecutive tissue sections. Kif5b and NCC were found to be expressed in the same tubular structures, indicating that Kif5b was expressed in the DCT.
The selective and asymmetric distribution of motor protein in mouse renal epithelial cells is not unique to Kif5b. KIFC3, a microtubule minus-end-directed motor, was reported to be specifically localized in distal tubules and loops of Henle, with strong staining in the apical area of the epithelial cells within these two nephron segments [26]. KIFC3 was also reported to transport annexin XIIIb-associated vesicles to the apical membrane, and this therefore implied that the plus-end-directed Kif5b was involved in basolateral transportation in epithelial cells in TAL and DCT, a process that may be essential for the normal physiological functions unique to these nephron segments such as transepithelial electrolyte transportation and regulation of urinary concentration [27].
Tubular cells in the TAL and the DCT are rich in mitochondria [28] and have a high density of Na + /K + -ATPase [29], which is essential for energy-demanding active intracellular transportation [30,31]. It was reported that Kif5b forms functional Kinesin-1 with Kinesin light chain-2 to regulate the transportation of the Na + /K + -ATPase-containing vesicle in alveolar epithelial cells [20]. We further analyzed the expression patterns of Na + /K + -ATPase in the kidneys of adult mice by double labelling of Na + /K + -ATPase and of markers of specific nephron segments ( Fig  5). It was found that Na + /K + -ATPase was basolaterally localized in the epithelial cells of THPpositive TAL and NCC-positive DCT (Fig 5D-5I), but not expressed in the LTL-positive PT or DAB (Dolichos biflorus agglutinin) positive CD (Fig 5A-5C and 5J-5L). Moreover, Kif5b and Na + /K + -ATPase were found to be co-localized in the same tubular structures (Fig 5M-5O), suggesting that Kif5b performs regulatory functions in Na + /K + -ATPase-containing vesicle transportation in renal epithelial cells of TAL and DCT that are similar to the functions that it performs in alveolar epithelial cells.
Basolaterally localized Na + /K + -ATPase can move sodium from the cell into the blood by consuming ATP, so as to establish and maintain an electrochemical gradient within the cell. Since Kif5b has been reported to transport mitochondria within cells [9,32], the basolateral distribution of Kif5b may facilitate the translocation of mitochondria to the vicinity of Na + / K + -ATPase where there is a high demand for ATP to sustain the transport of sodium out of the cell. In addition to Na + /K + -ATPase, there are also other transporters/receptors located on the basolateral membrane in TAL and DCT, which regulate sodium, calcium, and potassium re-absorption. By sensing the interstitial divalent mineral ion concentration, the basolaterally-localized Ca 2+ -sensing receptor (CaSR) [33] can regulate the transport efficiency of apical transporters: the Na-K-2Cl symporter (NKCC2) and K + channel in TAL, and the Na-Cl symporter (NCCT) in DCT [34]. The basolateral distribution of insulin-dependent (GLUT4) and insulin-independent glucose transporters (GLUT1) in the epithelial cells of TAL is important for electrolyte transport [35,36]; and in adipocytes, insulin-stimulated GLUT4 translocation is known to be mediated by Kif5b [15]. The involvement of Kif5b in the basolateral targeting of these receptors and/or transporters awaits further investigation. Nevertheless, it is possible that a deficiency of Kif5b in these tubules may make the re-absorption of sodium, potassium and chloride less efficient, thereby leading to renal disease similar to Bartter syndrome and/or Gitelman syndrome, which are electrolytic metabolic disturbance syndromes resulting from genetic mutations of ion channels such as NKCC2 in the TAL and NCCT in the DCT [37].

Kif5b Expression in Mouse Kidney
Kif5b fl/fl mice. Kif5b-Pax2KD mice died with limb muscle dystrophy at the perinatal stage, probably due to hypoxia and/or impairment of milk intake [23]. The Pax2-Cre activity has been proven to efficiently delete LoxP-flanked sequences in Pax2-expressing cells and their descendants [23,38]. Our analysis showed that positive immunostaining of β-galactosidase was observed in the ureteric bud and developing renal tubules of the kidneys of E19.5 R26R +/+ : Pax2-Cre embryos (Fig 6A), which overlapped with the location of expression of Kif5b. A decrease of approximately 90% in the level of Kif5b was observed in the kidneys of E16.5 [23] and E19.5 Kif5b-Pax2KD embryos (Fig 6B-6D). The overall histological morphology of kidneys was normal in E19.5 Kif5b-Pax2KD embryos (Fig 6E-6H). The three-layer arrangement of nephrogenic zone, cortex and medulla appeared to be as well organized in the mutant kidneys as in control kidneys. Thus, in the kidneys of the mutant mice, the cortex comprised both glomeruli and developing tubules (Fig 6E and 6F); and furthermore, the overall cellularity, asymmetry, size of the glomeruli in the mutant kidneys were comparable to those in control kidneys (Fig 6G and 6H). In the medullas of both groups of kidneys, the collecting ducts and loops of Henle formed rays in parallel from the cortex to the deep medulla, with the interstitial cells aligned perpendicularly to support these ducts (Fig 6G and 6H). No dilation or cyst formation was found (Fig 6E-6H). Taken together, no obvious abnormality in the morphogenesis of the glomeruli or the renal tubules could be detected by histological examination.
Dysregulation of cell proliferation and/or apoptosis results in renal dysplasia [39]. We examined the impact on cell proliferation of abolishing Kif5b-mediated intracellular transportation using BrdU labeling. As shown in Fig 7, BrdU incorporation was observed in metanephric mesenchyme cells, mesenchyme-derived epithelial structures and ureteric buds in the outer regions of kidneys, as well as interstitial cells located in the medulla region (Fig 7A). Quantitation of the number of BrdU + cells in ureteric bud branches (UB) and in metanephric mesenchyme (MM) revealed an approximately two-to three-fold diminution of cell proliferation in the kidneys of Kif5b-Pax2KD mice. By contrast, the cell proliferation rate was not affected in interstitial cells (IC) (Fig 7B). Based on TUNEL + analysis, there was no detectable increase in renal cell apoptotic rate (S2 Fig). In mice, pronephros (the predecessor of kidney) appears at E7.5 and develops to functional mesonephros at E9.5 and metanephros at E11 [2,3,40]. Afterwards, active nephrogenesis occurs to generate the mature kidney. The exact time at which Kif5b is knocked out in the kidneys of Kif5b-Pax2KD mice needs to be verified. However, Pax2-Cre starts to express at E9.5 [38], suggesting that Kif5b may be deleted as early as E9.5. Furthermore, expression of Kif5c mRNA in E12.5 kidneys (Fig 2) suggests that Kif5c may have a redundant function in replicating the role of Kif5b during kidney embryonic development and that this may contribute to the lack of obvious morphological abnormality in the kidneys of E19.5 mutant embryos.
In this study, we found that kidney-specific knock-down of Kif5b resulted in reduced cell proliferation rate in UB and MM. Similar phenotype was observed in Kif5b depleted pancreatic β cells, which resulted in reduced islet size in mutant mice [24]. Kif5b depleted Hela cell as well as MCF-7 cell also displayed growth inhibition [39]. Kif5b null mice showed midgestation lethality with severe growth retardation during 9.5-11.5 dpc [9]. Kif5b null embryos can survive the first several embryonic days indicating that cells can proliferate without Kif5b in specific cell types, but the proliferation rate was substantially decreased. However, primary cultures of Kif5b null myoblast cells had normal growth rate and cell cycle distribution pattern, indicating that Kif5b is dispensable for myoblast cell proliferation [23]. The specific function of Kif5b in cell proliferation is unclear up to now and it deserved further investigation.

Mislocalization of Na + /K + -ATPase in the kidneys of Kif5b knock-down embryos
The tubular localization of Kif5b in un-weaned mice, and the selective and asymmetric intracellular distribution of Kif5b both in the TAL and in the DCT in weaned mice, suggests that Kif5b is involved in physiological functions in renal tubules. Unfortunately, the early death of Kif5b-Pax2KD mice makes it impossible to undertake a functional analysis of Kif5b in the kidney. We analyzed the polarity of tubules by staining tissue sections with polarity markers (Figs 8 and 9). LTL, THP, NCC and DBA are apical markers of PT, TAL, DCT and CD, respectively. As in the kidneys of Kif5b fl/+ mice, all these markers were localized in the apical region of tubular epithelial cells in the mutant kidneys, indicating that the apical localizations of LTL, THP, NCC and DBA were not affected following the knock-down of Kif5b (Fig 8). However, the basolateral marker Na + /K + -ATPase was found to be present on the apical surface of tubular epithelial cells following the knock-down of Kif5b (Fig 9), suggesting that Kif5b is the motor protein involved in the transportation of Na + /K + -ATPase-containing vesicles in renal epithelial cells. Trejo et al. have previously reported that the traffic of the Na + /K + -ATPase depends on functional Kinesin-1 in the alveolar epithelium [20]. Here we provide direct evidence for the role of Kinesin-1 in the traffic of Na + /K + -ATPase in mouse kidney. We found that Kif5b co-localized with Na + /K + -ATPase, and knock-down of Kif5b depolarized Na + /K + -ATPase but did not affect the polarity of apical proteins. Since the Na + /K + -ATPase generates the electrochemical gradient that drives the movement of sodium from the cell into the blood [29], it could be suspected that genetic or epigenetic alteration in Kif5b (or Kinesin-1) may result in renal failure through depolarization of Na + /K + -ATPase, The physiological importance of Kif5b in mouse kidney is not clear and it remains uncertain whether Kif5b in the kidney is essential for mouse survival. In order to answer these questions, a more specific conditional knockout mouse model needs to be developed.
Supporting Information S1 Fig. (A) Amino acid sequence alignment for Kif5 member, 5-a, -b, -c. High variable sequences (underlined) in the neck region were selected as the antigens for Kif5b and Kif5c. Sequences (NGNATDINDNRSDLPC (992-1006aa)) located at the tail domain of Kif5a was selected as the antigen for Kif5a. (B) Purificaiton and testing the specificities of anti-Kif5a, 5b and 5c antibodies by western blot using mouse brain tissue lysate as the input. It is demonstrated that the anti-Kif5a (1:100), anti-Kif5b (1:1000) and anti-Kif5c (1:100) recognize protein bands of about 130, 120 and 110-120 kD respectively. Specific peptides can block the signals from the above antibodies in the Western blot. (C) The specificity for anti-Kif5b antibody has also validated on kidney tissue sections by immunostaining. Kif5b was strongly expressed in all developing tubules and ureteric bud in P0 kdineys (S1C Fig, upper left) and basolateral domain of some renal tubues in mature kidneys (S1C Fig, upper right). The positive staining was absent after preincubating the anti-Kif5b antibody with the 18-aa synthesized peptides (S1C Fig, bottom). Images are representative of kidney tissue sections from three mice. Scale bar = 50 μm.