Generation of Soluble Leptin Receptor by Ectodomain Shedding of Membrane-spanning Receptors in Vitro and in Vivo *

Leptin is an adipocyte-derived hormone with potent effects on food intake and body weight. Genetically obese rodents with mutations of leptin or leptin receptor develop morbid obesity and diabetes. The receptor for leptin, OB-R, is alternatively spliced to at least five transcripts, encoding receptors designated OB-Ra, -b, -c, -d, and -e. OB-Re does not encode a transmembrane domain and is secreted. In humans, transcripts corresponding to OB-Re have not been discovered. However, soluble leptin receptor does circulate in human plasma and represents the major leptin-binding activity. In this report, we attempted to determine whether the soluble leptin receptor may also be derived from membrane-spanning receptor isoforms by ectodomain shedding. Using stable cell lines expressing both OB-Ra, the most abundant leptin receptor isoform, and OB-Rb, the signaling form of the leptin receptor, we demonstrate that soluble leptin receptor protein can indeed be generated by proteolytic cleavage of these two receptor isoforms in vitro . Experiments using adenoviruses expressing dually tagged OB-Ra or Ob-Rb also demonstrate that soluble leptin receptor may be derived from ectodomain shedding of both receptor isoforms in vivo . Because our earlier and other studies have shown that the soluble receptors modulate the levels as well as activity of leptin, our findings suggest that regulated shedding

The leptin receptor, OB-R, is a member of the cytokine receptor family (5). It is encoded by the diabetes (db) gene, mutation of which also results in phenotypes similar to that exhibited by ob/ob mice. OB-R is alternatively spliced into at least five transcripts from a single gene. These transcripts encode proteins that are called the long (OB-Rb), short (OB-Ra, -c, and -d), and soluble (OB-Re) forms of the leptin receptor. With the exception of the soluble leptin receptor, other receptor isoforms differ from each other by the alternative use of a unique terminal coding exon (6). OB-Rb is essential in mediating leptin's weight-reducing effects via the hypothalamus (6,7).
OB-R is expressed in both the nervous system and in peripheral tissues. The relative expression levels of different receptor isoforms vary among tissues, possibly to allow leptin's biological activity to be more precisely regulated at various leptin target sites (8). OB-Rb is enriched in the hypothalamus, the site of leptin's action on food intake and body weight. Leptin activation of OB-Rb within this brain region results in the inhibition of neuropeptide Y/agouti-related protein neurons and activation of pro-opiomelanocortin neurons (9). Leptinactivated pro-opiomelanocortin neurons become depolarized and release anorexigenic peptides; leptin-inhibited neuropeptide Y/agouti-related protein neurons become hyperpolarized and reduce the release of orexigenic peptides (10). These neural circuits represent the main known downstream mediators of leptin's biological effect on food intake. OB-Rb can also activate signal transduction in a variety of peripheral tissues, including adipose tissue, T cells, endothelial cells, and pancreatic ␤-cells (11)(12)(13)(14)(15). More recently, it is also demonstrated that leptininduced fatty acid oxidation in muscle via 5Ј-AMP-activated protein kinase is mediated by both leptin acting on muscle directly and by functioning through the hypothalamic-sympathetic nervous system axis (16). Taken together, these results confirm that direct as well as indirect leptin signaling at these sites may be necessary for the many biological effects of leptin. In all cases, the presence of OB-Rb is required, as db/db mice that are without this receptor isoform do not respond to leptin.
Although OB-Rb is essential in mediating leptin's biological effects, other receptor isoforms may still be necessary for leptin to exert its full spectrum of in vivo functions. Among the short forms of the leptin receptor, OB-Ra is most abundantly expressed (17). It is enriched at the choroid plexus and brain microvessels, sites of the blood-cerebrospinal fluid barrier and the blood-brain barrier, suggesting that it may be involved in the transport of leptin across these barriers to reach the hypothalamus.
Previously, we have shown that the secreted form of the leptin receptor, OB-Re, circulates in mouse plasma and is capable of binding to leptin (18). In this report, we present evidence that circulating soluble leptin receptor may arise from both OB-Re mRNA and ectodomain shedding of membranespanning OB-R isoforms. We demonstrate that stable cell lines overexpressing OB-Ra or OB-Rb are capable of releasing soluble leptin receptor into the media by proteolytic cleavage, as detected by leptin-Sepharose pull-down assay. Furthermore, we demonstrate that adenovirus-mediated in vivo overexpression of OB-Ra or OB-Rb also results in release of soluble leptin receptor into plasma in mice via the same mechanism. Ectodo-main shedding of both OB-Ra and OB-Rb is detected by a monoclonal antibody recognizing the FLAG epitope tag fused in-frame with the amino terminus of recombinant receptor protein. Stable 293 cells expressing OB-Ra and OB-Rb release the soluble leptin receptor under serum-free conditions, suggesting that the cleavage of membrane-bound receptor may be mediated by proteases resident in the cell. Because the soluble leptin receptor affects leptin's effect on food intake and body weight in ob/ob mice (19), regulated ectodomain shedding of membrane-spanning leptin receptors may represent a novel mechanism of regulating leptin's biological activity.

EXPERIMENTAL PROCEDURES
Generation of Stable Cell Lines Overexpressing OB-Ra and OB-Rb-cDNA of Ob-Ra or Ob-Rb in the expression vector pcDNA3.1(Ϫ)/MycHisA was digested with Pme I and ligated into the same site of pcDNA5/FRT (Invitrogen). The resulting vectors containing OB-Ra or OB-Rb were co-transfected with pOG44 into a Flp-In TM host cell line (Invitrogen) using Fugene 6 Reagent (Roche Molecular Biochemicals) following the instructions by the respective manufacturers. Stable integrants were selected in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum and 100 g/ml hygromycin B (Clontech) for about 3 weeks. pcDNA5/FRT containing chloramphenicol acetyltransferase (CAT) 1 as insert was also transfected and selected as negative control in subsequent experiments.
Luciferase Reporter Analysis of Leptin Activation of OB-Ra and OB-Rb Cell Lines-Stable Flp-In cell lines expressing OB-Ra, OB-Rb, or CAT were seeded into 6-well plates until 50 -60% confluent. Cells were transfected with 1 g of p4xm67Luciferase plasmid (containing STAT3 response element and luciferase cDNA, kindly provided by Dr. Jim Darnell, the Rockefeller University). Twenty-two hours post-transfection, cells were washed and replaced with serum-free medium for 5 h, and leptin was added directly to a final concentration of 1 g/ml (50 nM). After 6 -24 h, cells were collected and luciferase activity measured using a luciferase assay system (Promega).

I-Leptin Binding to Stably Transfected Cells-Leptin binding to
Flip-In 293 cell lines was performed in 6-well plates as described (20). Briefly, cells stably transfected with CAT, OB-Ra, or OB-Rb were grown to about 90% confluence and washed with cold PBS. Cells were incubated with 60,000 cpm of 125 I-leptin (PerkinElmer Life Sciences) in the presence or absence of an excess of cold leptin (2 g/well) for 6 h at 4°C in a final volume of 1 ml of binding buffer (PBS containing 1% (w/v) bovine serum albumin (fraction V, Sigma)). At the end of incubation, unbound label was removed by two PBS washes; 1 ml of 1 N NaOH was added, and radioactivity in lysate was measured using a Canberra Model 2500 gamma counter.
Construction of Adenoviral Vectors Encoding Dually Tagged Leptin Receptors-The cDNA sequence encoding the signal peptide of OB-R was synthesized with XbaI and BamHI sites engineered at the 5Ј and 3Ј ends. Annealed oligonucleotides were ligated to pcDNA3.1(Ϫ)/MycHisA digested with the XbaI and BamHI to generate the vector pcDNA3.1(Ϫ)/MycHisA-SS. The cDNA encoding the long-form leptin receptor (OB-Rb) as well as the short-form receptor (OB-Ra) was PCRamplified from full-length expression vectors described earlier (21). The 5Ј-oligonucleotide contained the restriction site of BglII and the coding sequence of FLAG epitope (DYKDDDDK) as well as OB-R sequences immediately downstream of signal peptide. The 3Ј-oligonucleotide contains a HindIII site. PCR products of both OB-Ra and OB-Rb digested by BglII and HindIII were then ligated to pcDNA3.1(Ϫ)/MycHisA-SS digested with BamHI and HindIII. In this fashion, the resulting vectors contain a FLAG epitope that is inserted between the signal peptide and the remaining coding sequences of OB-R. The carboxyl terminus is already fused in frame with a c-myc epitope on the vector, as described earlier (21). Upon cleavage of the signal peptide during receptor translocation to the plasma membrane, the FLAG epitope becomes the amino terminus of recombinant OB-R protein, while the c-myc epitope becomes the carboxyl terminus. Final vectors of both OB-Ra and OB-Rb were sequence verified and transiently expressed in 293T cells to ensure that no cloning errors were introduced. Primer sequences are the following: Signal peptide forward (CL317), 5Ј-CTAGAAAGATGATGT-GTCAGAAATTCTATGTGGTTTTGTTACACTGGGAATTTCTTTATG-TGATAGCTGCACTTG-3Ј; reverse (CL318), 5Ј-GATCCAAGTGCAGCT-ATCACATAAAGAAATTCCCAGTGTAACAAAACCACATAGAATTTC-TGACACATCATCTTT-3Ј. FLAG (CL324), 5Ј-GAAGATCTGACTAC-AAGGACGACGATGACAAGAACCTGGCATATCCAATCTCT-3Ј (FLAG sequence is in bold and underlined), 3Ј end of OB-Ra (CL136), 5Ј-CCCAAGCTTAAGAGTGTCCGTTCTCTTTTG-3Ј, and 3Ј end of OB-Rb (CL135), 5Ј-CCCAAGCTTCACAGTTAAGTCACACATC-3Ј.
To generate adenoviruses expressing dually tagged OB-R, the entire coding cassette of OB-R in pcDNA3.1(Ϫ)/MycHisA constructed above was released by PmeI digestion and ligated to pSHUTTLE-CMV vector digested with EcoRV. Both shuttle vectors were first transfected into 293T cells to verify correct expression of encoded receptor protein (data not shown). Recombination with the adenovirus backbone vector pAdEasy1 was performed as described (22). Recombinant viruses were amplified and purified to a concentration of 10 12 virus particles/ml before being used for injection into mice. Adenovirus constructs encoding leptin (23) and ␤-galactosidase (24) were as described.
Animals and Adenovirus Infusion-C57Bl/6J mice at 9 weeks of age were purchased from the Jackson Laboratories (Bar Harbor, ME). These mice were divided into three groups and received control virus (AdCMV-␤-Gal), and receptor viruses encoding OB-Ra (AdCMV-OB-Ra), or OB-Rb (AdCMV-OB-Rb). After virus injection, mice were fed with standard rodent chow (Teklad, WI) with ad libitum access to both food and water. Mice were housed in specific pathogen-free barrier facilities and maintained on a 12-hour light/dark cycle. All procedures were performed in accordance with the policies of the Institutional Animal Care and Research Advisory Committee at the University of Texas Southwestern Medical Center. Viruses were injected into the jugular vein or tail vein of mice weighing between 25 and 30 grams. Each mouse received ϳ1 ϫ 10 11 total virus particles in 0.2 ml of PBS.
Plasma Preparation-Blood samples were collected from the tail vein into Eppendorf tubes coated with EDTA. Plasma was prepared by centrifugation (12,000 ϫ g, 5 min) and used in leptin pull-down assays to detect the soluble leptin receptor.
Northern Blot Analysis of Adenovirus-mediated OB-R Expression-Mice that received adenovirus injection were sacrificed under sodium pentobarbital anesthesia. Liver from each mouse was dissected immediately, washed with phosphate-buffered saline, and snap frozen in liquid nitrogen. Total RNA was extracted from tissues with Trizol reagent (Invitrogen). Northern blotting of OB-R was performed as described previously (17). Probe was labeled with [␣-32 P]dCTP (PerkinElmer Life Sciences) with a random primer labeling kit (PerkinElmer Life Sciences).
Soluble Leptin Receptor Assay-Plasma from AdCMV-OB-Ra or -Rb virus treated mice was diluted in PBS and incubated with leptin Sepharose resin overnight. After incubation, beads were washed with PBS three times and boiled in 2ϫ SDS sample buffer for 5 min. Resin suspension was loaded directly onto an 8% SDS-PAGE gel and blotted with an anti-leptin receptor polyclonal antibody as described (18). To determine the soluble leptin receptor in media of stable OB-R cell lines, tissue culture supernatant (with or without serum) was incubated with leptin-Sepharose resin overnight and blotted. 1 l of plasma sample from AdCMV-OB-Re-treated mice was run on SDS-PAGE directly to detect the levels of the soluble leptin receptor by Western blotting as described earlier (19).
Antibodies-Polyclonal antibody against the soluble leptin receptor was raised in rabbits against a synthetic peptide, EPLPKN-PFKNYDSK, corresponding to amino acid residues 145-58 of the extracellular domain of all OB-R isoforms, as described (18,19,21). This antibody was purified using an antigen column and is capable of recognizing the adenovirally produced soluble leptin receptor in plasma directly in Western blotting assays. When 293T cells are transiently transfected by vectors overexpressing OB-Ra or OB-Rb, expressed OB-R protein is readily immunoprecipitated with this antibody for detection with another antibody recognizing the c-Myc tag fused to the carboxyl terminus. Monoclonal antibodies against the FLAG epitope and the c-Myc epitope were from Sigma and Roche (Indianapolis, IN), respectively.

Generation of Stable Cell Lines
Overexpressing Membranespanning Receptors-To obtain permanent cell lines of OB-Ra and OB-Rb as reagents for leptin signal-transduction studies, we introduced cDNAs encoding OB-Ra, OB-Rb, or a control construct chloramphenicol acetyltransferase (CAT) into a 293 cell line using the Flp-In system. Expression of receptor protein was confirmed by immunoprecipitation of total cell lysate with a polyclonal antibody against the extracellular domain of the receptor, followed by Western blotting using a c-Myc monoclonal antibody that recognizes the carboxyl terminus that has been fused in-frame with the coding region of OB-Ra and OB-Rb (21). Fig. 1A shows the schematic diagram of receptor constructs used to transfect into 293 cells to generate stable cell lines. Both OB-Ra and OB-Rb protein can be detected by a c-Myc monoclonal antibody after immunoprecipitation (Fig.  1B). The amount of receptor protein expressed was much higher for OB-Ra than that of OB-Rb, in agreement with results from transient transfections (data not shown) and from other groups (20,25).
Functional Characterization of Stable OB-R Cell Lines-To test the functionality of OB-Ra and OB-Rb stable cell lines, we determined the abilities of leptin to bind to the cell surface of these cells as well as to activate STAT3 in these cells, using 125 I-leptin and luciferase reporter assay, respectively. Cells expressing CAT or OB-Ra failed to show leptin-induced STAT3 activation at all time points tested (data not shown). A robust response of luciferase activity was detected at both 6 h and 12 h after leptin addition to OB-Rb-expressing cells (Fig. 1C), demonstrating that a functional OB-R signal-transduction pathway is present in stably transfected OB-Rb cells. 125 I-leptin binding to the surface of OB-Ra or OB-Rb cell lines is about 8-or 4-fold higher than that of CAT cells (Fig. 1D). 125 I-leptin binding was essentially undetectable in the presence of an excess of cold leptin, demonstrating the specificity of this assay.

Release of Soluble Leptin Receptor into Medium by Stable 293
Cells Overexpressing Both OB-Ra and OB-Rb-Although the OB-Re transcript encoding soluble leptin receptor in rodents have been identified, such mRNA species in humans has not been found (26). Because soluble leptin receptor represents the main leptin-binding activity in human blood (27), an unresolved issue is how soluble leptin receptor is generated in humans. An earlier study reported that leptin-binding activity could be detected in the medium of cells transiently transfected with the human leptin receptor cDNA (20). This observation suggests that proteolytic cleavage of membrane-spanning leptin receptors could be the source of soluble leptin receptor in human circulation, analogous to that of the generation of the human soluble growth hormone receptor (28).
To test this possibility further, we determined whether the murine-soluble leptin receptor might also be generated by ectodomain shedding of membrane-spanning receptors, in addition to synthesis from the OB-Re transcript. Stable 293 cells overexpressing CAT, OB-Ra, or OB-Rb were incubated in the presence or absence of serum to determine whether soluble receptor immunoreactivity may be detected after leptin-Sepharose pull-down of culture supernatant and immunoblotting. Fig. 2 shows that a soluble leptin-binding protein with the size of OB-Re is readily detectable in the tissue culture supernatant of both serum-containing and serum-free medium of OB-Raexpressing cells. Signal is also detectable, although much weaker, from the supernatant of OB-Rb-expressing cells, con-

FIG. 1. Construction and characterization of stable 293 cells overexpressing OB-Ra or OB-Rb.
A, diagram of constructs generated. Both receptor proteins are fused with a c-Myc epitope tag at the carboxyl termini as described (21). The two shaded regions in the amino half of each receptor construct are sequences homologous to each other as well as to other cytokine receptors. TMR, transmembrane region. B, Western blotting of receptor protein expressed in OB-Ra or OB-Rb stable cell lines. Cells were lysed with radioimmune precipitation assay buffer, and recombinant OB-R was immunoprecipitated with a rabbit anti-receptor antibody, followed by detection with a monoclonal antibody recognizing the c-Myc epitope tag. C, luciferase reporter assay of leptin activation of OB-Rb. Flp-In 293 cells stably expressing OB-Rb were transfected with luciferase expression construct on a minimal promoter with four STAT3 DNA binding sites. Leptin (50 nM) was added for the duration indicated, and luciferase activity was determined. D, 125 I-leptin binding to stably transfected 293 cells overexpressing OB-Ra or OB-Rb. Cells were seeded in 6-well plates in quadruplicate, and labeled leptin was added to determine the number of receptor molecules present on the cell surface that are capable of binding leptin. Specificity of binding was confirmed by competition with an excess amount of cold leptin. sistent with its lower level of expression (Fig. 1B). In contrast, no signal was detected in lanes stably expressing CAT at all exposure levels (data not shown), suggesting that the soluble leptin receptor detected in the supernatant could not have come from that preexisting in the media. Interestingly, the soluble leptin receptor is also released into the media under serum-free conditions, indicating that the activity responsible may be a resident protease or proteases on the surface of these cells. Although the shed soluble leptin receptor is similar in size to that of OB-Ra (both run as ϳ150 kDa bands due to glycosylation), it is easily distinguishable from OB-Ra because it is not recognized by the anti c-Myc antibody that only recognizes the carboxyl terminus of OB-Ra (data not shown) and because OB-Ra, if not shed from plasma membrane, is an integral membrane protein and will not be present in the tissue culture supernatant, which is the starting material for leptinbead pull-down assay. In aggregate, these findings demonstrate that ectodomain shedding of membrane-spanning receptors may be another source of circulating soluble leptin receptor in rodents, in addition to that made from OB-Re mRNA. In addition, this result supports the hypothesis that a similar mechanism may also exist in humans, as has been reported recently (20).
Expression of Adenoviral Encoded OB-Ra and OB-Rb in Vivo-To determine whether the soluble leptin receptor may also be generated in vivo by ectodomain shedding, we generated dually tagged adenoviral vectors of OB-Ra and OB-Rb and expressed them in mice. This approach was taken because the size of the soluble leptin receptor released from stably transfected cells is indistinguishable from that produced by OB-Re cDNA (Fig. 2). To overcome this limitation, we inserted an amino-terminal FLAG epitope tag between the signal sequence of endogenous OB-R and the remaining coding sequences (Fig.  3A). If the soluble leptin receptor is cleaved from membranespanning receptor isoforms, it will be distinguishable from endogenous OB-Re by virtue of the presence of the FLAG tag, which is detectable by a monoclonal antibody.
We first verified the correct expression of dually tagged OB-R adenoviral vectors using 293 cells. 293 cells were infected with AdCMV-␤-Gal, AdCMV-OB-Ra, or AdCMV-OB-Rb. Cells were lysed with detergent-containing buffer. Virally expressed OB-R protein was immunoprecipitated with a polyclonal antibody against the extracellular region of the receptor, followed by sequential detection with monoclonal antibodies against the amino-terminal FLAG tag or the carboxyl-terminal c-Myc tag. Fig. 3B shows that bands of the same size were recognized by both antibodies, demonstrating that correct processing of both OB-Ra and OB-Rb has taken place.
Release into Plasma of Soluble Leptin Receptor by Ectodomain Shedding from Membrane-spanning Receptors-We pro- Leptin beads were washed with PBS three times and boiled directly in SDS sample buffer. Resin suspension was loaded onto an 8% SDS-PAGE gel and blotted with an anti-leptin receptor polyclonal antibody as described (18). In addition to the soluble leptin receptor protein with a size similar to that of OB-Re, a second band of smaller size (ϳ100 kDa) is also detected, which is likely the product of further proteolytic processing, because it is absent in supernatant of control cells, can bind leptin, and can react with the receptor antibody. The intensity of the smaller band also varies between experiments. Shed receptor is only visible from OB-Rb cells after longer expousure (right). Re, 1 l of plasma from a mouse infused with adenovirus overexpressing OB-Re, which is used as a positive control.

FIG. 3. Expression of dually tagged OB-R protein in 293 cells.
A, diagram of viral constructs generated. Adenoviruses encoding OB-Ra or OB-Rb were engineered to contain a FLAG epitope tag at the amino terminus immediately following the signal sequence. The carboxyl terminus is fused to a c-Myc tag. Mature OB-R protein on the cell surface should have the FLAG epitope at its amino terminus after removal of the signal peptide. B, detection of virally produced receptor protein with monoclonal antibodies against the amino-terminal FLAG tag or the carboxyl-terminal c-Myc tag. Total cellular extracts of infected 293 cells were prepared by lysis in radioimmune precipitation assay buffer, immunoprecipitated, and blotted using ECL reagents as for Fig. 2.

FIG. 4. Release of soluble leptin receptor into plasma via ectodomain shedding of membrane spanning receptor in vivo.
Dually tagged adenoviral expression vectors of OB-Ra (not shown) or OB-Rb were infused into C57Bl/6 mice and allowed to express OB-R protein from the liver. Plasma was obtained from control virus (Ad-CMV-␤-Gal)-injected mice or those that received OB-Rb. Leptin-Sepharose beads were incubated with plasma from each mouse as indicated on the figure and blotted with antibody recognizing the FLAG epitope tag, which is present only in soluble leptin receptor released from membrane-spanning full-length receptor (bottom). Viral expression of OB-R was demonstrated by Northern blotting from similarly loaded RNA samples of treated mice, using OB-R cDNA fragment as probe (top). ceeded to infuse adenoviruses encoding ␤-galactosidase, dually tagged AdCMV-OB-Ra or -OB-Rb into the tail vein of C57Bl/6 mice. Four days after virus administration, plasma was obtained from each mouse, and leptin beads were used to pull down soluble receptor protein present in plasma. Consistent with observations from Flp-In 293 cells stably transfected by OB-Ra or OB-Rb, soluble leptin receptor protein recognized by the amino-terminal FLAG antibody is detected in plasma of mice that received AdCMV-OB-Ra and AdCMV-OB-Rb but not in AdCMV-␤-Gal mice, demonstrating that the soluble leptin receptor may indeed be generated by ectodomain shedding of membrane-spanning receptor isoforms in vivo (Fig. 4). The cleavage is not leptin signaling-dependent because both OB-Ra and OB-Rb viruses are capable of releasing cleaved soluble leptin receptor (data not shown).
Taken together, our data demonstrate that the soluble leptin receptor may be generated by ectodomain shedding from membrane-spanning receptor isoforms both in vitro and in vivo. Because soluble leptin receptor stabilizes circulating leptin and affects its biological activity in vivo (19), controlled shedding of soluble leptin receptor under physiological conditions or dysregulation of this process under pathophysiological conditions may have important implications regulating leptin's weightreducing and other biological effects. DISCUSSION In our earlier studies, we demonstrated that the soluble leptin receptor circulates in plasma and is capable of binding to leptin (18). Our recently published results also demonstrated that the soluble receptor might play key roles in determining the amount of total leptin in circulation (19). Other studies also suggest that expression and plasma concentration of the soluble leptin receptor may be regulated (29,30). However, the origin of circulating soluble leptin receptor in plasma has not been clearly defined, especially as the mRNA splice variant encoding this receptor isoform in humans has not been found (26).
Because no post-translational modification of leptin occurs in vivo (31), soluble leptin receptor may be an important factor regulating leptin's availability and bioactivity. This hypothesis is supported by findings that in human plasma, soluble leptin receptor represents the major leptin binding activity (27).
The main site of expression of the mRNA splice variant encoding the soluble leptin receptor in vivo remains unknown. Previously, we failed to detect a signal for OB-Re when Northern blot analysis was performed (17). Available data have demonstrated that in mice, OB-Re is expressed by the placenta. Its expression starts at day 14 of pregnancy, peaking just before parturition to about 40-fold the level found in non-pregnant mice (29,32). In rats and humans, the pregnancy-associated rise of circulating leptin and its soluble receptor is relatively modest, achieving only a 2-fold increase versus a more than 40-fold increase in mice (33).
Our current study provided evidence of an additional mechanism for the generation of soluble leptin receptor by ectodomain shedding of membrane-spanning receptor, both in vitro and in vivo. This result is also in agreement with an earlier report on the generation of human soluble leptin receptor from transiently expressed membrane-spanning receptors in tissue culture (20).
Ectodomain shedding by proteolysis to yield soluble intercellular regulators has been observed for many proteins, such as tumor necrosis factor ␣ and transforming growth factor ␣ (34). The responsible protease for these proteins, tumor necrosis factor ␣-converting enzyme, is a member of the ADAM (a disintegrin and metalloproteinase) family of metalloproteinases. Molecules that are capable of inhibiting metalloproteinases are being tested as promising reagents for cancer therapy (35). At the present time, the modulation of the soluble leptin receptor on leptin biology is not well understood. Initial evidence of the presence of free and bound leptin in human circulation was obtained by gel-filtration chromatography of plasma containing added 125 I-leptin tracers (36). This and later studies showed that obesity is associated with decreasing levels of the circulating soluble leptin receptor in humans, whereas weight loss increases it (37)(38)(39)(40). In others studies, obese and normal weight individuals were found to have similar amount of total circulating soluble leptin receptor, yet in the obese most soluble leptin receptor is bound to leptin, whereas in lean individuals a much smaller percentage of soluble leptin receptor is bound to leptin (75% in the obese versus 33% in the lean); thus, the inability to up-regulate circulating soluble leptin receptor could be a factor in the pathogenesis of obesity (36,38,41). However, a larger percentage of leptin circulates in the free form in the obese (41). Opposite modifications of circulating levels of leptin and its soluble receptor also occur across the eating-disorder spectrum, with plasma levels of soluble leptin receptor significantly increased in patients with anorexia nervosa or bulimia nervosa, but decreased in patients with binge-eating disorder or those who are obese but are non-binge-eating (42). Soluble leptin receptor levels are also regulated by gender, adiposity, hormones, and rhLeptin administration (43). In aggregate, these studies suggest that the soluble leptin receptor may have important implications for the biological activity of leptin. Preliminary results of ongoing work in our laboratory using transgenic mice overexpressing the soluble leptin receptor support the hypothesis that increased soluble leptin receptor in wildtype mice is associated with decreased adiposity as well as decreased body weight. It is perceivable that future studies will shed more light on the usefulness of developing methods to regulate the generation of the soluble leptin receptor for therapeutic benefit.