INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

MOUSE Grb7 ("growth factor receptor bound") was the first identified member of an emerging family of adaptor signaling proteins (10). Grb7 encodes a 535-amino acid protein consisting of three regions, a carboxy-terminal src-homology 2 (SH2) domain, an amino-terminal proline-rich region, and a central region termed the GM domain (for Grb and Mig). The GM domain, which includes within it a pleckstrin homology (PH) domain (13), shows substantial conservation of sequence across a growing gene family that currently includes the following: mouse and human Grb7 (3, 10); mouse Grb10 (13) and the highly homologous human proteins Grb-IR (8), Grb-IR/Grb10 (3), and Grb10/IR-SV1 (12) ("IR" for insulin receptor binding, "SV" for splice variant); human Grb14 (1); and the Caenorhabditis elegans protein Mig10 (9).

The members of this family are cytosolic signaling proteins. These proteins possess no intrinsic enzymatic activity but instead function as adaptors, coupling events occurring at the cell surface receptor level with specific downstream signaling pathways. Critical for this adaptor function is the SH2 domain, a protein motif that binds phosphotyrosine in the context of the adjacent carboxy-terminal amino acids (5). Grb7 has been shown to bind, through its SH2 domain, with a number of tyrosine-phosphorylated receptor proteins including the following: epidermal growth factor (EGF) receptor (10); HER2/neu (erbB2) (22); Ret (15); and platelet-derived growth factor (PDGF) - and -receptors (26). To date, it has not been possible to identify either the subsequent signaling events or the physiological significance of these binding interactions in experimental systems.

Grb7 was cloned from a mouse embryonic cDNA library. Its gene expression has been demonstrated in kidneys in adult mouse (10) and human (3) tissue Northern blots. The mature and also the developing kidney comprise many specialized cells and structurally distinct nephron segments with heterogeneous, well-characterized functional phenotypes. Gene expression data in the mature and developing kidney and functional data in the developing kidney are available for a number of Grb7-binding growth factor receptors: HER2 is widely expressed along the nephron (6, 11, 17); Ret knockout mice are anephric (19); and PDGF -receptors are critical for normal glomerular development (21).

In an attempt to gain insight into the biological function(s) of the Grb7 protein, the following objectives were set: to confirm the presence of Grb7 protein and to determine the spatial pattern of Grb7 gene expression in the mature nephron; to investigate whether the Grb7 gene is expressed in the developing kidney; and if expressed, to determine the temporal and cell-specific patterns of its expression in the developing kidney. The patterns of Grb7 gene expression discovered are discussed with reference to the known expression patterns of Grb7-binding cell surface receptors. The findings were broadly consistent with Grb7 functioning in a basic signaling pathway in both developing and terminally differentiated epithelial structures.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Grb7 protein preparation and immunoblotting. Lysate from SKBR3 cells (a Grb7-overexpressing breast cancer cell line) was prepared as previously described (22). Proteins from SKBR3 cell lysate and rat kidney lysate (gift from J. M. Weinberg, University of Michigan) were resolved by SDS-PAGE, transferred to nitrocellulose, blocked in Tris-buffered saline (TBS)-BSA, and immunoblotted with anti-Grb7 rabbit polyclonal antibody (no. 188) (22), followed by horseradish peroxidase-conjugated protein A, then detection by the enhanced chemiluminescence system (ECL; Amersham, Arlington Heights, IL). Anti-Grb7 peptide antibody no. 188, was generated against amino acid sequence 264-279 (22).

Animals and dissection methods. Male Sprague-Dawley rats, 8 wk old, were anesthetized, and after interrupting aortic blood flow, the kidneys were perfused with 30 ml of cold saline followed by 30 ml of culture medium (DMEM; Sigma Chemical, St. Louis, MO) containing 1 mg/ml collagenase. Kidneys were removed, cut into slices, and incubated in the DMEM-collagenase solution for 22 min at 37°C. Microdissection was performed at 4°C under a stereomicroscope, and lengths of dissected segments were measured with an eyepiece micrometer. In general, 10 glomeruli or 6-10 mm of tubule segments were dissected and pooled to constitute one sample. Samples were placed in 100 µl guanidine isothiocyanate buffer (GITC buffer: 4 mol/l guanidine isothiocyanate, 25 mmol/l sodium acetate, pH 6.0, and 0.8% -mercaptoethanol), snap frozen in liquid nitrogen, and stored at 80°C. Strips of outer cortex, outer medulla, and inner medulla, obtained from both rat and adult CD-1 mouse kidneys, were also placed in GITC buffer for RNA extraction. Pregnant CD-1 mice were killed at different gestational ages, and embryos were aseptically removed, placed in ice-cold culture media (equal volumes DMEM and Ham's F-12 medium), and transferred individually to cold PBS (pH 7.4), then metanephric organs consisting of ureteric bud and undifferentiated mesenchyme were removed. For RT-PCR purposes, whole kidneys from embryonic day 11 (E11) embryos or portions of kidney from E14 and E17 were placed in 100 µl GITC buffer, snap frozen in liquid nitrogen, and stored at 80°C. For in situ hybridization, whole mouse embryos and metanephric organs were fixed for 12 h in 4% paraformaldehyde, cryoprotected in 20% sucrose, embedded in optimal cutting temperature compound (OCT), and stored at 80°C.

RT-PCR: RNA isolation and reverse transcription. RNA from the above preparations was isolated using a modification of the CsCl technique as previously described (4). Briefly samples were thawed in an ice slurry and sonicated for 15 s; 20 µg of ribosomal RNA from Escherichia coli (Boehringer Mannheim, Indianapolis, IN) was added to each sample; samples (in 100 µl of GITC buffer) were layered onto a gradient of cesium chloride (100 µl of 97% and 20 µl of 40% cesium chloride in 25 mmol/l sodium acetate buffer) in a 250-µl polycarbonate ultracentrifuge tube and centrifuged for 2 h at 300,000 g in an ultracentrifuge (model TLA 100; Beckman Instruments, Fullerton, CA) with a fixed-angle rotor. The RNA pellet was redissolved in 0.3 mol/l sodium acetate and ethanol precipitated.

Reverse transcription was performed in the presence of 100 IU Moloney murine leukemia virus reverse transcriptase (RT) (Superscript; BRL, Gaithersburg, MD), 0.5 µg oligo(dT)_12-18 (Pharmacia, Piscataway, NJ), 20 IU RNasin (Promega Biotech, Madison, WI), 10 mmol/l dithiothreitol, 0.5 mmol/l dNTP (Pharmacia), and 1% BSA (Boehringer Mannheim) in the buffer provided by the manufacturer in a total volume of 20 µl. Prior to the addition of RT, dNTPs, and BSA, the reaction mixture was incubated at 65°C for 5 min to allow the primers to anneal to the poly-A tail of mRNA. cDNA was synthesized at 42°C for 1 h, then precipitated with 1 µl of linear acrylamide, 4 M ammonium acetate, and 100% ethanol. The pellets were redissolved in Tris-EDTA buffer at a dilution adjusted so that each 2 µl of cDNA corresponded to 1 mm of tubule or 1 glomerulus.

RT-PCR: polymerase chain reaction. Primers were chosen from the Grb7 mouse cDNA sequence to amplify a product of 310 bp, which would be cut once by Xho I into two fragments of sizes 133 and 177 bp, respectively. The sequence of sense primer was 5'-TATGGACTTCTCTGGCCATG-3' (bp 1499-1518) and that of the antisense primer was 5'-TGACTTTCTGCAGATGGCAC-3' (bp 1790-1809). To control for variations in RNA amount and efficiency of reverse transcription, PCR amplification for -actin was also performed. The sequences of the -actin primers were as follows: sense 5'-AACCGCGAGAAGATGACCCAGATCATGTTT-3' (bp 384-413), and antisense 5'-AGCAGCCGTGGCCATCTCTTGCTCGAAGTC-3' (bp 705-734).

Reactions were performed in a total volume of 50 µl in the presence of 5 pmol of each oligonucleotide primer, 200 µmol/l dNTP, 10 mmol/l dithiothreitol, 50 mmol/l KCl, 1.5 mmol/l MgCl₂, 10 mmol/l Tris · HCl, pH 8.3, 0.001% gelatin, 1.25 IU of AmpliTaq DNA polymerase (Perkin-Elmer Cetus, Norwalk, CT), and 1.5 µCi [³²P]dCTP (Amersham). The samples were first denatured at 94°C for 3.5 min. The PCR cycle was programmed as follows: 94°C for 40 s, 58°C for 40 s, and 72°C for 40 s. PCR was run for 30 cycles (except for microdissected nephron segments where 33 cycles were run), and the last cycle was followed by additional incubation of 8 min at 72°C. Negative PCR controls (no added cDNA) were run with each experiment. PCR products were subjected to size separation by PAGE. The band intensity was determined by phosphorimaging with the Phosphor Analyst software on a GS-250 Molecular Imager System (Bio-Rad, Hercules, CA). The identities of the PCR products were determined by restriction digestion with Xho I at 37°C for 1 h in the buffer provided by the manufacturer.

Generation of cRNA probes and in situ hybridization. Bluescript SK-Grb7 (305-1905) and Bluescript KS-Grb7 (305-1905) were linearized with Xba I and EcoR I, respectively, to permit generation of 1,600 bp antisense and sense probes as follows: 1 µg of linearized plasmid DNA, 2 µl of 10× digoxigenin-RNA labeling mix (Boehringer), 2 µl of 10× transcription buffer (Boehringer), and 2 µl of T3 RNA polymerase (Boehringer) were added to a microcentrifuge tube at 4°C (final volume 20 µl); incubation for 2 h at 37°C was followed by the addition of 2 µl of RNase-free DNase I (Boehringer), with further incubation at 37°C for 15 min; finally 2 µl of 0.2 M EDTA, pH 8.0, was added to stop the reaction. Unincorporated ribonucleotides were separated using Sephadex G50 spin columns. Aliquots of each purified probe were run on a 1% TBE (Tris-borate, EDTA) minigel, with the remainder being precipitated with 1/10th volume 3 M sodium acetate and 2 vol of 100% ethanol at 80°C, centrifuged at 14,000 rpm for 15 min, and washed with 95% ethanol at 4°C. Probes were subjected to limited alkaline hydrolysis for 35 min at 60°C in 50 µl of an RNase-free solution of 40 mM NaHCO₃ and 60 mM Na₂CO₃, pH 10.2, to produce a probe size of ~0.2 kb before finally being reprecipitated, washed, and resuspended in a solution of 98 µl diethyl pyrocarbonate (DEPC)-treated distilled water and 2 µl RNase inhibitor (Boehringer). Labeling efficiency of the probes was checked by dot blots. [³³P]UTP-labeled Grb7 probes were also generated and used following previously described methods (2).

Steps in the nonradioactive hybridization procedure were carried out at room temperature except where otherwise indicated. Cryosections of 8 µm thickness were cut and thaw-mounted onto poly-L-lysine-coated Superfrost Plus slides (Fisher). Slides were dried at room temperature, fixed in 4% paraformaldehyde in 1× PBS for 20 min; washed sequentially (5 min each) in 3× PBS, 1× PBS, and 1× PBS; dehydrated in a graded ethanol series 30-100%, 2 min each; air dried and then stored in a box with desiccant at 20°C. At the time of hybridization, slides were warmed to room temperature, rehydrated in a graded ethanol series, washed in PBS, and then hybridization was performed by the following steps: fixation in 4% paraformaldehyde for 20 min; washing three times in DEPC-treated PBS, 10 min each; acetylation for 10 min in autoclaved 0.1 M triethanolamine containing 0.25% acetic anhydride; rinsing in DEPC-treated distilled H₂O and dehydration as before through a graded ethanol series.

Probe stocks of sense and antisense probe were prepared by adding the desired volume of probe (2 µl of 100 ng/µl probe stock per slide) and RNase-free tRNA (1 µl of 1 ng/µl stock per each 100 µl of hybridization stock solution) to the desired volume of hybridization stock solution (100 µl per slide). The hybridization stock solution contained 50% deionized formamide, 1 mM Tris · HCl, pH 7.4, 30 mM NaCl, 6× Denhardt's solution, and 10% dextran sulfate. Hybridization was performed under sealed coverslips in a humid chamber at 60°C for 16 h overnight. Posthybridization steps were as follows: removal of coverslips in 2× SSC; further 5 min in fresh 2× SSC; 30-min treatment in 0.5 M NaCl, 10 mM Tris, pH 7.4, and 0.01 mg/ml RNase A at 37°C; sequential washing in 2× SSC 15 min, 1× SSC (15 min), 0.5× SSC (60 min), and finally 0.5× SSC (5 min at room temperature). Next slides are rinsed in PBS with 0.3% Tween 20 (PBST) and blocked by incubating in filtered PBST/1% BSA for 30 min. Anti-digoxigenin-alkaline phosphatase conjugate (Boehringer Mannheim) was diluted 1:1,000 in PBST/BSA and applied to the slides for a 1-h incubation in a humid chamber at room temperature. After removal of antibody by washing twice (30 min each) in PBS, signal was detected by incubation with nitro blue tetrazolium and 5-bromo-4-chloro-3-indolyl phosphate in chromogenic buffer (100 mM Tris · HCl, pH 9.5, 100 mM NaCl, and 50 mM MgCl₂). Incubation was continued for 24-48 h in a humid chamber. Slides were counterstained with Methyl Green 1%, rinsed in water, dehydrated through graded ethanol, and covered with mounting medium.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Grb7 protein and mRNA expression in adult kidney. Prior to initiating gene expression studies, we documented the presence of Grb7 protein in mature rat kidney homogenate (Fig. 1A). The presence of Grb7 mRNA in cortex and medulla of both adult rat and mouse kidneys was next defined by RT-PCR (Fig. 1B), and this was a consistent finding over six repeated experiments designed to optimize the PCR conditions. Restriction digestion with Xho I and size resolution of the amplification product and its fragments was used to confirm origin from Grb7 cDNA (Fig. 1C). Because the genomic sequence of Grb7 has not yet been determined, we cannot confirm that our primers flank an intron-exon boundary. However, for all sets of total RNA isolated/cDNA generated, at least one experiment on a negative control sample (no reverse transcriptase) was performed to check for genomic contamination. Figure 1D shows parallel PCR reactions using Grb7 primers on samples generated from total RNA, isolated from a single experimental animal, and respectively processed with or without reverse transcriptase. No 310-bp Grb7 fragment could be detected in the non-reverse transcriptase-treated control sample, although a small amount of amplification of a larger fragment was seen.

View larger version (37K): [in this window] [in a new window]   Fig. 1.   Grb7 mRNA and protein expression in the adult kidney. A: immunoblot analysis using rabbit polyclonal anti-Grb7 antibody (no. 188) (Ref. 22) on 20 µg of protein from the Grb7-overexpressing SKBR3 breast cancer cell line (lane 1) and rat kidney homogenate (lane 2). Detection was by enhanced chemiluminescence system (ECL, Amersham), 1-min exposure. Anti-Grb7 peptide antibody no. 188 stains a number of nonspecific bands that can be eliminated by preimmunoprecipitation with a second polyclonal anti-Grb7 antibody (no. 222) (Ref. 22). B: Grb7 cDNA is generated by reverse transcription of total mRNA from adult mouse (lanes 1 and 2) and rat (lanes 3 and 4) kidney cortex and medulla. A product of the correct predicted size, 310 bp, is amplified. C: identical PCR product and restriction digest fragment sizes (Xho I) amplified from total mouse cDNA (lanes 3 and 4) and from a positive plasmid control (lanes 1 and 2) containing the full-length Grb7 cDNA sequence. D: RT+/RT, plus or minus reverse transcriptase. In RT samples, there was no Grb7 310-bp product.

Grb7 mRNA expression in microdissected nephron segments. Three experiments with cDNA separately generated from microdissected adult rat nephron segments were performed. The results of a representative experiment are shown in Fig. 2. Grb7 was present in all tubular segments assayed but was absent from glomeruli or arcuate artery specimens in each of the performed experiments. The -actin amplification accompanying each experiment confirmed that variations in RNA amount and efficiency of reverse transcription were not responsible for the absence of detectable Grb7 mRNA in the glomerular or arcuate artery samples.

View larger version (47K): [in this window] [in a new window]   Fig. 2.   Grb7 mRNA expression (top; with -actin at bottom) in microdissected nephron segments. ARC ART, arcuate artery; GLOM, glomerulus; PCT, proximal convoluted tubule; PST, proximal straight tubule; TDL, thin descending limb; MTAL, medullary thick ascending limb; CTAL, cortical thick ascending limb; CCD, cortical collecting duct; OMCD, outer medullary collecting duct; IMCD, inner medullary collecting duct.

Grb7 mRNA in embryonic tissues. Next, the presence of Grb7 gene product was demonstrated by RT-PCR in mouse embryonic kidneys from days E11, E14, and E17. Three experiments confirmed detectable Grb7 throughout this critical embryonic period. In the first two experiments, there was more detectable Grb7 at E17 than at E11, despite there being less baseline -actin in the E17 cDNA samples. A semiquantitative experiment was then performed (Fig. 3), by determining carefully the linear range for -actin and Grb7 amplification for cDNA samples generated from developing kidney at days E11, E14, and E17 of development. Again, a moderate increase in Grb7 gene expression (normalized for -actin) was seen between E11 and E17 (Fig. 3). More Grb7 product is seen at identical dilutions of the cDNA samples when E17 is compared with E11. By contrast less -actin product is seen at identical dilutions of the cDNA samples when E17 is compared with E11.

View larger version (53K): [in this window] [in a new window]   Fig. 3.   Grb7 mRNA (top; with -actin at bottom) in mouse kidney at embryonic days E11, E14, and E17. PCR product counts are indicated above each lane, adjusted volume counts ×103/mm2 (Phosphor Analyst). Dilutions are indicated below each lane. Separate linear regression lines were generated by plotting log PCR product counts (detected by the phosphorimager) against log serial dilutions of cDNA for Grb7 and -actin at each stage (E11, E14, and E17). Average r2 = 0.99, and the average of the six slopes was 1.09 ± 0.1, indicating near-linear range Grb7 (and -actin) amplification. Assuming linear amplification, all -actin product counts were changed to adjusted counts for a standard sample dilution of 1:10,000, and averages were taken for these standardized counts at E11, E14, and E17. The ratio between these averaged -actin standardized product counts at E11, E14, and E17 was 1:15:0.34. Similarly, the ratio of averaged Grb7 standardized product counts (all counts changed to adjusted counts for a standard dilution 1:100) at E11, E14, and E17 was 1:26.25:1.81. After normalizing for baseline differences in -actin between the cDNA samples, the ratio of Grb7 product detected in this experiment was 1:1.8:5.3 at E11, E14, and E17, respectively.

Following an initial in situ hybridization experiment with radioactive probes, three subsequent in situ hybridization experiments (multiple sections in each, from E14 embryos and E17 and newborn kidneys) were performed with digoxigenin probes. Sense probes were included in all experiments. The Grb7 digoxigenin sense probes gave essentially no background staining and were therefore preferred over the radioactive sense probes, which gave background staining. The patterns of Grb7 mRNA illustrated in Fig. 4, A-F, and Fig. 5, A and B, are representative of repeated findings in all experiments. Grb7 gene product was absent from undifferentiated mesenchymal tissue but diffusely present in epithelial structures derived from both ureteric bud and mesenchyme (Fig. 4, A-F). Less dense staining of podocyte-destined cells, in the bilayered epithelium at the lower poles of the S-shaped bodies (Fig. 4C), and of early glomerular structures (Fig. 4B) was a consistent feature. In addition to these results, in situ hybridization (with both digoxigenin and radioactive probes) in E14 embryos also defined epithelial expression of Grb7 in the gut and lung (Fig. 5, A and B). Although the radioactive probes suggested the presence of Grb7 mRNA over background in the liver, this was not detected by digoxigenin probes.

View larger version (127K): [in this window] [in a new window]   Fig. 4.   Cellular localization of Grb7 mRNA in mouse embryonic kidney. A: negative control experiment using sense riboprobe in E17 kidney. B: diffuse pattern of gene expression in renal epithelial structures in E17 kidney. C: Grb7 mRNA detected in epithelia of ureteric bud (3 arrows at top) and mesenchymal origin in E14 kidney. Note relatively decreased staining in podocyte precursor cells of the bilayered epithelium at the lower pole of S-shaped body (broken arrow, bottom left). D: detectable Grb7 mRNA is not limited to the tips of ureteric bud (E14). E and F: an experiment with a Grb7 antisense radioactive riboprobe shows similar findings (E14).

View larger version (144K): [in this window] [in a new window]   Fig. 5.   Epithelial specific Grb7 expression (day E14) in the developing lung (A) and gut (B).

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

In the design of the above experiments, the three objectives proposed in the introduction have been addressed. For the first time, we have documented actual Grb7 protein in kidney tissue. RT-PCR provided a very sensitive tool with which to detect Grb7 gene expression in microdissected nephron segments and in the metanephric kidney from the earliest stages of its development. In the developing metanephric kidney, the major focus of this work was to determine temporal and cell-specific patterns of Grb7 expression rather than to quantitate gene expression. The dramatic changes in the relative predominance of different mesenchymal and epithelial cell types during metanephrogenesis make the interpretation of quantitative gene expression studies based on whole organ cDNA samples difficult. These same dramatic changes emphasize the need for cellular localization of gene expression. For these reasons, we switched from RT-PCR approaches to in situ hybridization techniques to complete the characterization of Grb7 expression during metanephrogenesis.

Grb7 has been found to be coamplified, overexpressed, and bound to HER2 in breast cancer tissue and in breast cancer cell lines (22). The overexpression of HER2, a member of the EGF receptor family of growth factor receptors, portends a worse prognosis in node-positive breast cancer (20). Increased expression of the HER2 protooncogene has also been reported in other human adenocarcinomas including gastrointestinal, prostate, and renal carcinomas (18, 20, 24, 25). In transgenic mice, expression of an activated allele of HER2 resulted in preneoplastic lesions in the kidneys and lungs and scattered transgene expression elsewhere was correlated with epithelial hyperplasia (23). The physiological expression of HER2 seems to correlate with the expression pattern we have observed for Grb7. HER2 has been shown to be normally expressed on human fetal epithelial cells by immunohistochemistry (11, 17). The expression is strongest in the kidney, gut, and liver. Within the kidney, expression has generally appeared high in epithelial structures of mesenchymal and ureteric bud origin, low in the parietal epithelium of Bowman's capsule, and undetectable in glomeruli and in undifferentiated mesenchyme. A gene knockout of c-erbB2 (HER2) in mice was embryonic lethal at embryonic day E10.5, with cardiac and neural developmental anomalies (7) preventing study of the role of this gene in epithelial cell development. As with Grb7, persistent expression of HER2 in the adult nephron has been detected by immunohistochemistry (6, 17). These similar results for Grb7 and HER2 expression suggest that the functional effects mediated through the Grb7-Her2 interaction must have significance for both a rapidly growing, branching, and dividing epithelial cell phenotype (as seen in development or adenocarcinoma) and for a mature, terminally differentiated cell phenotype.

A phosphotyrosine-dependent association between Grb7 and the Ret tyrosine kinase has been demonstrated in vitro and in vivo (15). In gene knockout mice, the Ret / phenotype is characterized by renal agenesis and defects in the development of the enteric nervous system (19). The cellular localization of Grb7 gene expression includes the tips of the ureteric bud branches, exclusive sites for Ret expression during kidney development (14). In addition, Grb7 mRNA is present in the developing mouse kidney as early as E11. This temporal and spatial pattern of Grb7 expression is consistent with the possibility that Grb7 is one of the intracellular mediators of the critical inductive signaling that occurs through stimulation of the Ret receptor. Grb7 expression in the developing gut was limited to the epithelial lining and does not correlate with the restricted sites of Ret expression in gut wall neural connections (14).

A Grb7/PDGF receptor interaction (26) gives rise to the hypothesis that Grb7 gene expression might be important in those cells destined to become the glomerular mesangium. We have found Grb7 gene expression to be undetectable by RT-PCR and in situ hybridization in mature glomeruli and barely detectable by in situ hybridization in primitive glomeruli, and therefore we cannot provide evidence to support this hypothesis. Interactions between Grb7 and other receptor tyrosine kinases, such as Met, may also be functionally important in vivo in the developing kidney.

A full understanding of the biological sequelae of signaling through the Grb7 protein family remains elusive. Grb10 isoforms bind strongly to the insulin receptor, but their role in insulin signaling has not been determined. Grb10 is expressed in a different set of tissues than is Grb7 (13). The C. elegans related gene mig-10 is involved in neuronal cell migration, but the signaling pathway on which it sits is also unclear (9). The SH2 and PH domains are considered to be involved in membrane targeting because they promote interactions with activated receptor tyrosine kinases (defined) or membrane phospholipids (not defined). Recently, it was suggested that a sequence of ~100 conserved amino acids within the GM domain of the Grb7 protein family, immediately amino terminal to the PH domain, might serve as a Ras-binding domain (16). The provocative observation was based on alignments with known and putative Ras-binding proteins. Grb7 proteins, however, did not score very high in this alignment. We have tested this observation with several experimental designs and found no detectable Grb7/G-protein interactions (unpublished data).

In summary, although the downstream signaling events relayed through Grb7 remain enigmatic, it is probable on the basis of these observations that Grb7 serves important physiological functions in growth factor signaling in terminally differentiated epithelia along the nephron and in developing epithelia in the kidney, lung, and gut.