Renal FcRn Reclaims Albumin but Facilitates Elimination of IgG

Renal FcRn Reclaims Albumin but FacilitatesElimination of IgG

Menaka Sarav, Ying Wang, Bradley K. Hack, Anthony Chang, Mark Jensen, Lihua Bao, andRichard J. Quigg

Section of Nephrology, Department of Medicine, The University of Chicago, Chicago, Illinois

ABSTRACTThe widely distributed neonatal Fc receptor (FcRn) contributes to maintaining serum levels of albuminand IgG in adults. In the kidney, FcRn is expressed on the podocytes and the brush border of theproximal tubular epithelium. Here, we evaluated the role of renal FcRn in albumin and IgG metabolism.Compared with wild-type controls, FcRn�/� mice had a lower t1⁄2 for albumin (28.7 versus 39.9 h) and IgG(29.5 versus 66.1 h). Renal loss of albumin could account for the former, suggested by the progressivedevelopment of hypoalbuminemia in wild-type mice transplanted with FcRn-deficient kidneys. Further-more, serum albumin levels returned to normal in FcRn�/� recipients of wild-type kidneys after removingthe native FcRn-deficient kidneys. In contrast, renal loss could not account for the enhanced eliminationof IgG in FcRn�/� mice. These mice had minimal urinary excretion of native and labeled IgG, whichincreased to wild-type levels in FcRn�/� recipients of a single FcRn-sufficient kidney (t1⁄2 of IgG was21.7 h). Taken together, these data suggest that renal FcRn reclaims albumin, thereby maintaining theserum concentration of albumin, but facilitates the loss of IgG from plasma protein pools.

J Am Soc Nephrol 20: 1941–1952, 2009. doi: 10.1681/ASN.2008090976

The neonatal Fc receptor (FcRn) is a heterodimerconsisting of a MHC class I-like heavy chain and�2-microglobulin light chain.1 It is so named be-cause it transports maternal IgG across the placentaand the fetal small intestine to passively immunizethe fetus.2,3 FcRn is also widely expressed in adults,including in vascular endothelium, liver, spleen,lung, and kidney;4 – 6 hence, its gene designation asIgG Fc receptor, alpha chain transporter (Fcgrt).

FcRn binds to albumin and IgG with high affin-ity at acidic pH (�6.5) but not at physiologic pH(7.4).3,7,8 Notably, the interactions of FcRn with al-bumin are distinct from those with IgG.9 Informa-tive studies by the Ward group using cultured en-dothelial cells have shown that FcRn acts onpinocytosed IgG, saving it from lysosomal degrada-tion and instead recycling it intact into the medi-um.10 Consistent with its important role in albuminand IgG metabolism, mice with targeted mutationsof FcRn or its �2-microglobulin partner have lowerserum t1⁄2 for albumin and IgG compared with wild-type mice.11,12 Taken together, it is thought that vas-

cular endothelial FcRn plays a critical role in con-serving intravascular albumin and IgG in the wholeanimal.

The main function of the kidney is to filterplasma and subsequently retain the vast majority ofthat fitrate. The glomerular filtration barrier ismade up of three components—the fenestrated en-dothelium, the glomerular basement membrane,and the podocyte slit diaphragm—together func-tioning as a size- and charge-selective barrier.13 Theconcept that this barrier prevents serum proteinssuch as albumin and IgG from appearing in anyappreciable quantity in the urine has been called

Received September 15, 2008. Accepted April 23, 2009.

Published online ahead of print. Publication date available atwww.jasn.org.

Correspondence: Dr. Richard Quigg, Section of Nephrology, TheUniversity of Chicago, 5841 South Maryland Avenue, MC5100,AMB-S523, Chicago, IL 60637. Phone: �1-773-702-0757; Fax:�1-773-702-4816; E-mail: [email protected]

Copyright � 2009 by the American Society of Nephrology

BASIC RESEARCH www.jasn.org

J Am Soc Nephrol 20: 1941–1952, 2009 ISSN : 1046-6673/2009-1941 1941

into question with recent evidence for active tubular uptake ofalbumin.14 In the kidney, FcRn is expressed on the podocytesand the brush border of the proximal tubular epithelial cell(PTEC),6 with both sites clearly relevant to IgG metabo-lism.15,16 To further define the roles for renal and extrarenalFcRn in albumin and IgG metabolism, we used fluorescence-labeled proteins and the transplantation of kidneys betweenmice differing only by a targeted FcRn gene knockout.

RESULTS

Clearance of Albumin and IgG by Renal FcRnThe serum concentrations of Alexa-labeled albumin and IgGover time after intravenous injection were determined in wild-type and FcRn�/� mice. As expected,11,12 greater quantities ofalbumin and IgG remained intravascular from 24 to 96 hpostinjection in wild-type mice compared with FcRn�/� ani-mals (Figure 1, A and B). The second-phase t1⁄2 of albumin was39.9 � 2.2 h in wild-type mice and 28.7 � 2.6 h in FcRn�/�

mice (P � 0.071), whereas that of IgG was 66.1 � 12.0 h inwild-type mice and 29.5 � 3.9 h in FcRn�/� mice (P � 0.040).Interestingly, FcRn�/� mice had a nearly 3-fold lower volumeof distribution (Vd) for IgG compared with wild-type animals,as evidenced by the higher serum quantities within 12 h ofinjection; extrapolated y-intercepts (time “zero”) were19,442 � 944 and 6432 � 337 U/ml in FcRn�/� and wild-typemice, respectively (P � 0.042). Consistent with the quickerdisappearance of IgG from the circulation, the curves in thetwo strains crossed early after injection (Figure 1B).

The fractional urinary excretion (FE) of albumin and IgG (rel-ative to creatinine) 3 h after intravenous injection was consider-ably higher than subsequent values, which were fairly constant ina given animal. These suggested there was saturation initially fol-lowed by attainment of steady state. As such, data were dividedinto those 3 h and 24 to 96 h postinjection, which illustrated thatFEalbumin was modestly higher in FcRn�/� mice compared withwild-type mice (Figure 1C), whereas FcRn�/� mice had markedlylower FEIgG compared with wild-type animals (Figure 1D). Thus,although urinary losses could account, at least in part, for thereduced t1⁄2 of albumin in FcRn�/� mice, this was not a viableexplanation in the case of IgG.

Three complementary approaches were utilized to extendupon these quantitative data and examine potential degradationof urinary albumin and IgG. Using trichloroacetic acid, Alexa 488and Alexa 594 fluorescence intensities were consistently over 96%precipitable, support that Alexa probes remained protein-boundin sera and urines from these short-term experiments. Next,Sephacryl S-100 size-exclusion chromatography to separate uri-nary albumin was performed as described by the Comper labora-tory.17 This showed albumin appeared as a single intact peak inurines from both mouse strains (Figure 1E). Albumin and IgGwere also labeled with infrared dyes to allow direct visualization oftheir urinary forms after separation by SDS-PAGE. By this ap-proach, albumin and IgG were also predominantly intact (Figure

1F). These data collectively showed that FcRn�/� mice excretedgreater amounts of urinary albumin but less IgG than wild-typemice after systemic administration. Moreover, albumin and IgGappeared intact in urines, thereby excluding significant proteoly-sis.

Intrarenal Localization of Albumin and IgGThe fate of intravenous-injected, Alexa-labeled albumin andIgG in kidneys was determined quantitatively (Figure 2A) andqualitatively 5 to 30 min postinjection in wild-type (Figure 2,B, D, F, G, I, and J) and FcRn�/� mice (Figure 2, C, E, H, andK). In both strains, Alexa 594-albumin associated with PTECsby 5 min postintravenous injection (Figure 2, B and C). Inwild-type kidneys, this declined rapidly along with the appear-ance of intracellular vesicular bodies containing Alexa 594-albumin (Figure 2J, arrows). In contrast, Alexa 594-albuminremained associated with the brush border of FcRn�/� PTECs,with lesser amounts reaching intracellular compartments (Fig-ure 2K, arrows). Greater quantities of Alexa 488-IgG were inthe kidneys of wild-type mice compared with FcRn�/� animals(Figure 2A). In both mouse strains, Alexa 488-IgG was associ-ated with PTECs. In wild-type mice, this was in basolateralcompartments largely distinct from Alexa 594-albumin (Fig-ure 2, F and I), whereas it appeared to be associated togetherwith Alexa 594-albumin in the brush border of FcRn�/� kid-neys (Figure 2K, arrows). The representative photomicro-graphs shown in Figure 2 illustrate a variation in PTEC uptakeof Alexa probes, which was highly consistent in all experimentsbut was also clearly not related to relative location within thekidney (i.e., cortical versus medullary). Alexa 488-IgG was alsoevident within wild-type (Figure 2J) but not FcRn�/� glomer-uli (Figure 2K); in the former, this appeared to be mesangialand in association with podocyte cell bodies marked with anti-synaptopodin antibodies (Figure 2J, arrows).

Kidney Transplantation to Determine Roles for Renaland Extrarenal FcRnTo evaluate albumin and IgG handling by renal and extrarenalFcRn in greater detail, we transplanted kidneys between wild-type C57BL/6 mice and those in which the FcRn targeted mu-tation was carried forward in 13 backcrosses onto the C57BL/6strain (Table 1). In all animals, a single native nephrectomywas performed at the time of transplantation; thus, in thesemice, there were single functional native and transplanted kid-neys. In studies to determine whether a FcRn-sufficient kidneycould “rescue” a FcRn-deficient animal, a second native ne-phrectomy was performed after several weeks, thereby gener-ating a FcRn�/� mouse with a single functional wild-type kid-ney. Evidence that the transplant protocol itself did not affectthe function of FcRn-sufficient or deficient kidneys includednormal renal function (Table 1) and absence of renal pathol-ogy in the transplanted kidneys at sacrifice; specifically, no kid-ney had glomerular or vascular pathology, and only 1 of the 18transplanted kidneys had tubulointerstitial pathology (mildinterstitial nephritis, scored as 1 of 4).


1942 Journal of the American Society of Nephrology J Am Soc Nephrol 20: 1941–1952, 2009

Albumin Metabolism after Renal TransplantationAs designed, the experimental protocol allowed us to examinethe fate of native albumin over time after the various transplantcombinations. In unmanipulated wild-type and FcRn�/�

mice, serum albumin levels were 30.0 � 2.1 and 26.9 � 3.8

mg/ml, respectively (NS). Control experiments were per-formed in which a wild-type kidney was transplanted into awild-type mouse, thereby leading to single functioning (FcRn-sufficient) native and transplanted kidneys. In these animals,serum albumin levels remained constant over time (Figure 3A,

Figure 1. Role of FcRn to metabolize intravenously injected Alexa-labeled albumin and IgG. (A and C) Alexa 594-labeled albumin and (B andD) Alexa 488-labeled IgG were injected into wild-type (n � 3) and FcRn�/� mice (n � 4) for (A and B) measurements of sera t1⁄2 and (C andD) urinary excretion. For clearance data (A and B), logarithmically transformed data were used from individual animals (r � 0.940, median0.988) and were averaged to create the curves shown in the figures. All data as presented were normally distributed (Anderson–Darling tests,H� � 0.1) and are means with SEM values shown in C and D. �P � 0.005 versus all other groups (ANOVA followed by Fisher’s pairwisecomparisons). Urines from wild-type and FcRn�/� mice were also assessed by (E) Sephacryl S-100 HR size-exclusion chromatography and (F)infrared-imaged SDS-PAGE gels after injection of unlabeled albumin or IRDye-labeled albumin and IgG, respectively. In F, wild-type mouseurine before injection was used as control, the entire gels are shown (i.e., extending beyond the dye fronts), and the locations of labeledproteins, including IgG heavy (H) and light (L) chains and molecular weight markers are indicated.

BASIC RESEARCHwww.jasn.org

J Am Soc Nephrol 20: 1941–1952, 2009 Albumin and IgG Handling by Renal FcRn 1943

E), supporting that the presence of a transplanted wild-typekidney, or the transplant protocol itself, did not alter albuminmetabolism. FcRn�/� mice receiving a wild-type kidney hadincreasing albumin levels over time. This rose further to nor-mal when the remaining FcRn-deficient kidney was removedin these mice (Figure 3A, E) (P � 0.02), evidence to supportthe ability of an FcRn-bearing kidney to reclaim albumin. Incontrast, the transplantation of a single FcRn-deficient kidney

into a wild-type host resulted in progressively lower serumalbumin levels over time (Figure 3A, E). Furthermore, theseanimals developed anasarca, thus exhibiting a clinical featureof the nephrotic syndrome. The explanation for why FcRn�/�

mice have relatively preserved serum albumin levels and lackany evidence for nephrotic syndrome includes increased he-patic albumin synthesis,23 presumably developing over timebecause the gene deficiency is present at conception in these

Figure 2. Intrarenal appearance of intravenous-injected Alexa 594-albumin and Alexa 488-IgG. Renal Alexa 594-labeled albumin (red)and Alexa 488-IgG (green) were quantified (A) over time and visualized (B and C) 5, (D and E) 15, and (F through K) 30 min afterintravenous injection into (C, E, H, and K) FcRn�/� and (B, D, F, G, I, J) wild-type mice. Stained blue in these photomicrographs is (F,J, and K) wheat germ aggluttin or (I) the podocyte marker synaptopodin. The arrows depict Alexa 594-albumin in PTECs. (K) In FcRn�/�

mice, albumin colocalized with Alexa 488-IgG, whereas in wild-type mice this was in distinct vesicular structures (J). Alexa 488-IgG waspresent within (F, arrowheads) basolateral aspects of tubules, (I and J) glomeruli of wild-type mice, and (I, arrowheads) in regionsassociated with synaptopodin-bearing podocytes. (A) Nonparametric data (Anderson–Darling tests, H� � 0.006) were derived asdescribed in Concise Methods from one animal per time (4 to 12 measurements each) and presented as medians. The dashed line fordata of IgG intensity in wild-type mice is shown simply to draw attention to the break in the y-axis. Differences between wild-type andFcRn�/� mice were determined using Kruskal–Wallis testing for albumin (P � 0.001) and IgG (P � 0.011). Original magnifications were200� in B through H, 400� in J and K, and 600� in I.



mice. Our transplant protocol evidently did not allow suchcompensation to occur in wild-type mice.

To further evaluate albumin metabolism in FcRn�/� micewith a FcRn-sufficient wild-type kidney, we evaluated the t1⁄2 ofintravenous-injected Alexa 594-albumin. Data from four indi-vidual FcRn�/� mice with a transplanted wild-type kidney areshown (Figure 3B, gray symbols and lines), with average datafrom FcRn�/� (red line) and control transplanted wild-typemice (blue line) as reference. These data show the t1⁄2 of albu-min was normalized in FcRn�/� mice containing a FcRn-suf-ficient wild-type kidney. To confirm that the transplant proto-col did not affect urinary albumin metabolism, experimentsevaluating urinary albumin by size-exclusion chromatography(i.e., comparable to those shown in Figure 1E) showed albuminappeared as a single peak (data not shown). Taken together,these results are entirely consistent with the premise FcRn inthe kidney is important to salvage albumin, such that an FcRn-deficient kidney “leaks” albumin, whereas a wild-type kidneyretains it.

Additional experiments were performed to determine po-tential effects of ischemia-reperfusion injury or uninephrec-tomy on albumin metabolism in FcRn�/� mice. Within 3 h ofreperfusion, urinary albumin was significantly increased inwild-type and FcRn�/� mice (876 � 166 and 887 � 45 �g/mgcreatinine, respectively), and returned to baseline values by day7. With this, serum albumin levels did decrease to a mild ex-tent, reaching 86.5 � 3.7 and 86.6 � 7.6% of baseline values inwild-type and FcRn�/� mice, respectively, 1 to 2 wk afterreperfusion. In FcRn�/� mice subjected to uninephrectomy,urinary albumin did not change and serum albumin levels ei-ther remained stable or rose slightly, consistent with thepremise that kidney FcRn is important in albumin metabo-lism.

IgG Metabolism after Renal TransplantationSerum IgG levels were 10.9 � 1.7 and 6.2 � 1.2 mg/ml in acohort of unmanipulated wild-type and FcRn�/� mice (n � 7to 8) (P � 0.038). Although mice remained housed in a barrierfacility, the move from a specific pathogen-free facility and asystemic inflammatory response postsurgery were likely to ac-count for a mild elevation in serum IgG in all transplant groupsafter surgery. Thus, IgG levels were 14.3 � 1.8 and 16.6 � 1.6mg/ml in wild-type mice 4 wk after transplantation of a wild-type or FcRn�/� kidney, respectively. Moreover, IgG levels in

Figure 3. Kidney FcRn normalizes serum albumin levels and t1⁄2 inFcRn�/� mice. (A) Shown are serum albumin levels over time inwild-type recipients of wild-type or FcRn�/� kidneys, andFcRn�/� recipients of wild-type kidneys. Also shown are datafrom after removal of the second native kidney in FcRn�/� recip-ients of a wild-type kidney, such that a sole wild-type kidneyremained (sera samples collected at this time are indicated by thearrow). Data are mean � SEM; in the latter point, the SEM of 0.2mg/ml was within the symbol. (B) After intravenous injection,Alexa 594-labeled albumin was measured over time in FcRn�/�

recipients of a transplanted wild-type kidney after both nativeFcRn�/� kidneys were removed. Individual animal data areshown, along with curves fitted to data. Also shown for referenceare averaged data from FcRn�/� mice, as well as wild-type recip-ients of wild-type kidney transplants. All data were normally dis-tributed (Anderson–Darling tests, H� � 0.1); that the groups weredifferent was confirmed by ANOVA (P � 0.016).

Table 1. Transplant groups

Group Donor Kidney Recipient Mouse BUN (mg/dl)a n

WT 3 FcRn�/� WT kidney FcRn-deficient host 31.6 � 3.0 8b

FcRn�/� 3 WT FcRn-deficient kidney WT host 27.3 � 0.8 7WT 3 WT WT kidney WT host 30.3 � 3.2 3All recipients had a single native nephrectomy performed at the time of transplantation. WT, wild type.aBUN values 4 wk after transplantation (mean � SEM).bThe second native FcRn-deficient kidney was removed 4 wk or later after transplantation.



FcRn�/� mice with a transplanted wild-type kidney were11.9 � 1.4 mg/ml 4 wk posttransplant and rose further to13.0 � 2.8 mg/ml 10 d after removal of the remaining nativeFcRn-deficient kidney.

To better understand the role for renal FcRn in IgG metab-olism, the fate of intravenous-injected Alexa 488-IgG was stud-ied in transplanted animals. Controls were wild-type mice intowhich wild-type kidneys were transplanted, which had compa-rable IgG clearance as wild-type mice (Figure 4A, averagesshown as a blue line). Interestingly, the addition of a FcRn-deficient kidney to a wild-type host did not affect the t1⁄2 of IgG

(Figure 4A, magenta symbols and lines for individual mice). Incontrast, the already low t1⁄2 of IgG in FcRn�/� mice (Figure 4A,red line showing averages) was further reduced when they hada sole functioning wild-type kidney (Figure 4A, charcoal sym-bols and lines for individual mice) (i.e., 29.5 � 3.9 to 21.7 �1.3 h).

Renal handling of Alexa 488-IgG was also determined inthese studies by measuring FEIgG 48 h after intravenous injec-tion (Figure 4B). Individual FEIgG measurements from wild-type mice receiving wild-type and FcRn�/� kidneys, andFcRn�/� mice with wild-type kidney transplants are shown,together with data from untransplanted FcRn�/� mice forcomparison. For these studies, FcRn�/� mice with wild-typekidney transplants had both native FcRn-deficient kidneys re-moved before study. In an additional animal studied in whicha single native kidney was left in place, FEIgG was 5.6% at 48 h.These data show the FEIgG depended strictly on the presence ofa FcRn-sufficient kidney; in any circumstance in which therewas at least one functional wild-type kidney, FEIgG was in-creased considerably beyond that evident in FcRn�/� mice,indicating IgG was excreted through FcRn-sufficient kid-ney(s).

Localization of Albumin and IgG in Transplanted andNative KidneysThe data presented above suggested that kidneys can salvagealbumin through PTEC FcRn; yet, renal FcRn also appears toserve as a conduit for IgG excretion, which is magnified whenthe host animal lacks FcRn in all other sites, including endo-thelia. Our transplant studies allowed us to examine the fates ofAlexa-labeled albumin and IgG, which was particularly infor-mative in those instances in which FcRn-sufficient and FcRn-deficient kidneys were present in wild-type and FcRn animals.Control studies included wild-type mice with wild-type kidneytransplants, which showed findings in kidneys comparable towild-type mice without transplantation (Figure 5, A and B).

As was true of kidneys in FcRn�/� mice that did not un-dergo transplantation, FcRn-deficient kidneys transplantedinto wild-type mice (or native to transplanted FcRn�/� ani-mals) had significant quantities of Alexa 594-albumin in PTECbrush borders 3h after intravenous injection (Figure 5C). Al-bumin was also present in PTEC brush borders of wild-typekidneys in wild-type (Figure 5, A and B) or FcRn�/� hosts(Figure 5, E and F), particularly in early postglomerular (i.e.,S1) segments (Figure 5F, arrow). As observed previously, therewas evidence for PTEC uptake in the wild-type kidneys (Figure5B, arrow).

Wild-type kidneys transplanted into wild-type hosts hadAlexa 488-IgG in the tubulointerstitium, basolateral aspects oftubules, and glomeruli (Figure 5, A and B); this pattern wasaccentuated in wild-type kidneys in a FcRn�/� host, particu-larly in interstitial spaces between glomeruli and tubules insites of microvascular endothelium (Figure 5, E through G,arrowheads). As with native FcRn�/� kidneys, FcRn-deficientkidneys in a wild-type host had lesser amounts of IgG in tu-

Figure 4. Urinary excretion of IgG requires renal FcRn. Alexa488-labeled IgG was injected into wild-type recipients of wild-type or FcRn�/� kidneys, and FcRn�/� recipients of wild-typekidneys after both native kidneys had been removed for measure-ments of (A) sera t1⁄2 and (B) urinary excretion. Individual animaldata are shown. For reference, data of nontransplanted FcRn�/�

mice (n � 4), derived as detailed in the text and legend to Figure1, are also shown. Data in both graphs were normally distributed(Anderson–Darling tests, H� � 0.1). (A) Curves were fitted to datafrom individual animals (transplants between wild-type andFcRn�/� mice) or averaged data (nontransplanted FcRn�/� mice,wild-type to wild-type transplants); the latter for ease of visualiza-tion. Data were first analyzed by ANOVA (P � 0.011) followed byFisher’s pairwise comparisons. *P � 0.05 versus wild-type recip-ients of wild-type and FcRn�/� kidneys. (B) Data were first ana-lyzed by ANOVA (P � 0.036) followed by Fisher’s pairwise com-parisons. *P � 0.05 versus other groups.



bules (Figure 5C), and little within glomeruli (Figure 5D). Ofnote was the presence of apparent aggregates of Alexa 488-IgGin tubulointerstitium (Figure 5D, asterisk) and glomeruli (notshown). Taking all available data together, these support theconcept that podocyte FcRn promotes IgG clearance throughthe glomerulus, which is partly reclaimed by the PTEC using itsFcRn; when absent, IgG accumulates within renal sites. In ad-dition to these sites involved in maintenance of the glomerularfiltration barrier and urinary IgG, it also appears that renalvascular FcRn plays an important role in the kidney’s handlingof IgG.

DISCUSSION

The results from these studies are consistent with past work byseveral investigators showing that FcRn is a key protein in-volved in systemic albumin and IgG metabolism.4,9,12,15,16,18 –23

Yet we have extended these by transplanting kidneys betweenFcRn-sufficient and FcRn-deficient mice and monitoring sera,

urines, and kidney tissue sites for fluorescence-labeled albu-min and IgG. These results show FcRn differs in its roles insideand outside of the kidney, as well as in the handling of albuminand IgG. Thus, as predicted, FcRn within the kidneys is respon-sible for reclaiming albumin and restoring it to the systemiccirculation.6 Yet, surprisingly, renal FcRn facilitates the loss ofIgG from plasma protein pools. Furthermore, extrarenal FcRn,presumably that in endothelial cells, affects albumin and IgG inthe systemic circulation and its accessibility to the kidney.

The quantity of glomerular filtrate is remarkable, approxi-mating 50 plasma volumes daily. Nearly all of the water andmany other filtered molecules must be actively returned to thecirculation for survival. The glomerular filtration barrier haslong been considered a passive size- and charge-selective sieve,capable of restricting IgG and albumin on the basis of size andcharge, respectively.13,24 Still, this view has not lacked its de-tractors.25–27 A recent study using contemporary imaging tech-niques showed the normal glomerular ultrafiltrate containedsubstantially higher quantities of albumin than expected; thiswas reclaimed by normal but not diseased PTECs.14 Candi-

Figure 5. Intrarenal appearance of intravenous-injected Alexa 488-IgG and Alexa 594-albumin in transplanted kidneys. Shown arerepresentative images from (A and B) wild-type kidneys and (C and D) FcRn-deficient kidneys in a wild-type host and wild-type kidneysin a FcRn�/� host (E through G) 3 h after intravenous injections. (B, D, and F) Stained blue is the podocyte marker synaptopodin.Albumin (labeled red) was prominent in brush border of tubular cells in FcRn-deficient kidneys in (C and D) a wild-type host as well asin (F, arrow) brush borders of early PTECs and in (B, arrow) vesicular structures of wild-type kidneys. IgG (labeled green) was presentin wild-type kidneys in FcRn�/� hosts in (F) glomeruli and (E through G, arrowheads) tubulointerstitial spaces, whereas lesser amountswere present in FcRn�/� kidneys in wild-type hosts, some of which were in aggregates (D, asterisk). Original magnifications were 200�for A, C, and E; 600� for B, D, and F; and 800� for G.



dates for albumin-binding proteins in the PTEC brush borderinclude megalin and cubilin.28 From the brush border of thePTEC, albumin can enter a high capacity transcytotic path-way29 –32 or a lower capacity pathway directed toward lysoso-mal delivery and degradation.17,32–34

Our studies have confirmed and extended upon the studiesof Anderson and colleagues relating to albumin metabolism inFcRn�/� mice,12,23 with the t1⁄2 for albumin in FcRn�/� mice75% that of wild-type mice. The wild-type and FcRn�/� micewe studied were at steady state, with net production equalingthat lost through metabolism. Reduced albumin levels inFcRn�/� mice have been shown to reflect the net of greaterproduction and clearance.23 Although we did not directlystudy production, our data showing 28% greater clearance ofalbumin in FcRn�/� mice (i.e., measured albumin t1⁄2 values of28.7 and 39.9 h) are consistent with previous data showing32% greater clearance observed in FcRn�/� mice (in dimen-sionless units, not permitting direct comparisons).23 It shouldbe noted that steady-state albumin levels we measured in wild-type and FcRn�/� mice were 30.0 and 26.9 mg/ml, respectively,which does differ in magnitude from previous data (respectivevalues of 36.3 and 21.0 mg/ml23), which we cannot explain.Intrarenal handling of albumin was different between the twostrains. In such a setting, there were greater quantities of albu-min associated with PTEC brush borders in FcRn�/� mice,whereas wild-type mice appeared to have uptake of albumininto intracellular compartments followed by its disappearance.The Alexa 594 label did remain protein-bound in sera andurines, excluding metabolism to peptides; this was confirmedin short-term experiments by three complementary tech-niques. Beyond that, the precise fate of this intracellular albu-min was undefined in our studies. Taking all data together, itdoes seem likely that a substantial portion was returned to thesystemic circulation.

A key role for intrarenal FcRn in albumin metabolism wasconfirmed by transplanting kidneys between FcRn-sufficientand -deficient animals. The mild hypoalbuminemia present inFcRn�/� mice was partially corrected when there was a FcRn-sufficient kidney and fully restored when the second FcRn�/�

kidney was removed. Moreover, placing a FcRn-deficient kid-ney in a wild-type host led to prompt development of hy-poalbuminemia and features of the nephrotic syndrome.

Overall, it does seem likely that the modest increase inFEalbumin we observed here in FcRn�/� mice is relevant andaccountable for the development of hypoalbuminemia withtime. PTEC FcRn may not bind urinary albumin directly butappears necessary to protect it from being lost in the urineand/or taken up and degraded by the PTEC. The data from theComper group do indicate the albumin reclamation processesin the kidney operates at or near saturation.14,17,30 Our data inwild-type and FcRn�/� mice at steady state given a bolus ofalbumin (as Alexa-labeled albumin) are consistent with thisconcept; within hours of administration, the ability to reclaimalbumin is overcome, leading to the high values of FEalbumin inboth groups. This fits well with the concept that glomerular

filtered albumin is much higher than previously thought14 andthat albumin reclamation occurs in the proximal tubules, assupported, for instance, in data from protein overload mod-els.30 The intrarenal appearance of Alexa-labeled albumin wasalso different between the two groups, with very little observedwithin 5 min in wild-type tubules. Later on, there was evidencefor vesicular trafficking of albumin within both groups, whichwe feel is consistent with the low-capacity/high-affinity meta-bolic pathway. FEalbumin quickly comes back to much lowerlevels, presumably as the amount of albumin once again reach-ing the tubules declines to that able to be handled. Taken to-gether with the data of others, we feel our results are consistentwith an important role for renal tubular FcRn in the high-capacity/low-affinity reclamation pathway.

Our experimental design also allowed comparison betweenanimals bearing a FcRn-sufficient and a FcRn-deficient kidney.There were remarkable differences in serum albumin levelsdepending on whether the host was a wild-type or FcRn�/�

mouse; wild-type mice with FcRn-deficient kidneys had pro-gressive hypoalbuminemia, whereas FcRn�/� mice with FcRn-sufficient kidneys increased albumin levels. This undoubtedlyreflects the intricacies of albumin metabolism in the FcRn�/�

mouse, including greater hepatic synthesis and altered endo-thelial cell handling.23

Cultured human PTECs have a capacity for bidirectionalIgG transport;15 in vivo, basolateral to apical transport ofIgG would increase the urinary appearance of IgG. Thatpodocytes use their FcRn6,35 to clear the glomerular filtra-tion barrier was recently shown by Akilesh and Shaw in theirstudies of FcRn�/� mice in which injected heterologous IgGaccumulated over time in the subepithelial space.16 Addingcomplexity to IgG clearance is its propensity to form latticedimmune complexes containing antigen and complementcomponents. This is clearly relevant in disease states such aspostinfectious GN36 and membranous nephropathy,37 inwhich there is the development of IgG aggregates largeenough to be visible histologically. Complement receptor 1in human podocytes (and complement factor H in ro-dents)38,39 has the potential to interact with C3b in immunecomplexes. In addition to FcRn and complement receptor 1,the podocyte has other features of an immunologically ac-tive cell, including CD80 (B7-1),40 CD2-associated pro-tein,41 and the ability to present antigen in vitro.42 Whetherthe podocyte, with its repertoire of complement and Fc re-ceptors, actually serves as an antigen-presenting cell in vivocomparable to dendritic cells,43,44 or whether it is involvedin a multistep process of antigen presentation ultimatelyrelying upon parenchymal dendritic cells (for which there isample supply in the kidneys45), as true for FcRn-bearingintestinal epithelial cells,18 is an exciting prospect to workout in future studies.

Renal FcRn is responsible for urinary clearance of IgG asshown by the very low FEIgG in FcRn�/� mice, which wasmarkedly increased by providing a single transplanted wild-type kidney. Clearance through the podocyte is supported cir-



cumstantially by the presence of Alexa 488-labeled IgG in glo-meruli (and near podocytes) in wild-type but not FcRn�/�

mice. It does seem unlikely transport through glomeruli ac-counts for the large differences in renal clearance of IgG be-tween FcRn-deficient and -sufficient strains. In wild-type kid-neys, there was considerable IgG in the tubulointerstitium,supporting vasa recta endothelium transported IgG. As withpodocyte-trafficked IgG, it is intriguing to speculate that theintrarenal dendritic cell network could also survey IgG (in im-mune complexes) to potentially generate an immune response.The subsequent fate of IgG could include retrieval throughlymphatics and transport by PTECs in a basolateral to apicalroute. Overall, the latter is favored by our data showing local-ization of IgG in PTEC basolateral sites, as well as the urinarydisposition of IgG.

Endothelial cells are believed to nonspecifically take upalbumin and IgG from the circulation. Subsequently, FcRnprotects both from a catabolic fate, and instead they arerecycled to the circulation or transcytosed into the intersti-tial space.4,10 The differences in IgG Vd between FcRn�/�

and wild-type mice approximates the difference betweenplasma and extracellular fluid volumes, supporting that theFcRn in endothelium does promote IgG transcytosis to theinterstitium. This is further supported by examining the fateof IgG in FcRn�/� mice with a wild-type kidney, in whichthere was appearance of sizable quantities of IgG and anextremely low t1⁄2. This can be explained by a combination ofIgG restricted to the intravascular space outside of the kid-ney because of a lack of endothelial FcRn, and a conduit forits urinary excretion in the normal FcRn-containing kidney.It is important to add that in this circumstance, the renallymphatics were no longer intact because of the transplan-tation protocol, so recycling through renal lymph was notpossible. On balance, the low t1⁄2 in FcRn�/� mice with orwithout a transplanted wild-type kidney also must includeits loss outside the kidney, for which endothelial cell catab-olism is the likeliest candidate.4,10,18

In summary, we have shown a role for renal FcRn in albu-min and IgG metabolism. The locations of FcRn in the kidneyinclude in podocytes, PTECs, and vascular endothelia, eachappearing to have individualized functions. Albumin is filteredat the glomerulus and reabsorbed by the PTEC. FcRn in thelatter is necessary for reclaiming albumin, such that in its ab-sence, there is progressive loss of albumin through catabolismand/or excretion in the urine. Podocyte FcRn facilitates clear-ance of IgG that has traversed the glomerular filtration barrier,whereas renal vascular endothelial FcRn removes IgG from thecirculation into the interstitium. Once there, the fate of IgGincludes uptake by the PTEC at its basolateral site, followed bytranscytosis into the urine. It appears the endothelial cell bedkeeps albumin within the intravascular space and moves IgGtranscellularly to access extracellular spaces. These functionsare logical because they maintain oncotic and other intravas-cular functions of albumin while providing IgG to underlyingtissue as necessary for immunity to infectious agents.

CONCISE METHODS

MiceFcRn�/� mice (Fcgrttm1dcr)11 were obtained from Jackson Laborato-

ries (Bar Harbor, ME). Mice were backcrossed onto the C57BL/6

background for 13 generations or more. Mice used in these studies

were derived from intercrosses between FcRn�/� mice. Age-matched

wild-type C57BL/6J control mice were also purchased from Jackson

Laboratories. Initial experimentation to determine albumin and IgG

clearances and renal localizations utilized normal wild-type and

FcRn�/� mice (n � 30). All work with mice was performed under the

auspices and approval of the University of Chicago Animal Care and

Use Committee.

Renal TransplantationIn these studies kidney transplantation was performed in mice to

evaluate renal and extrarenal effects of FcRn (Table 1), comparable

to our past studies with TNF receptor 1,46 Toll-like receptor 4,47

complement receptor 1-related gene/protein y,48 and complement

factor H.39 Male mice between 8 to 10 wk of age were used either as

a kidney donor or recipient. Blood and urine were collected from

all animals before transplant surgery to establish baseline values.

Donor mice were anesthetized, and the donor left kidney was re-

moved with artery, vein, and ureter attached and was preserved in

cold saline on ice. The recipient was then anesthetized, and the left

kidney was excised. Renal transplantation was performed with

end-to-side anastomoses of the donor renal vein, artery, and ureter

to the recipient inferior vena cava, aorta, and bladder, respectively.

Total cold ischemic time of the donor kidney ranged between 45

and 60 min. Blood and urine were collected weekly posttransplan-

tation. At 4 wk posttransplantation, FcRn�/� mice with wild-type

kidneys underwent nephrectomy of the right native kidney, such

that they relied solely on the transplanted FcRn-sufficient kidney

to provide renal function. At the time of sacrifice, native and trans-

planted kidneys were processed for histopathological analyses and

scored by a renal pathologist (A.C.) for GN, glomerulosclerosis,

interstitial nephritis, tubular atrophy, and arteritis, each on a scale

from 0 to 4.

In separate experiments to determine whether reperfusion injury

and/or the presence of a solitary kidney could affect albumin metab-

olism, wild-type and FcRn�/� mice (n � 3 each) had the left renal

pedicle clamped for 30 min or underwent left uninephrectomy.

Thereafter, serum albumin levels were measured biweekly.

Measurements of Albumin and IgG ClearancesPurified mouse albumin (Innovative Research, Novi, MI) and total

IgG (Lampire Biologic Laboratories, Piersville, PA) were respec-

tively labeled with Alexa 594 and 488 fluorescence probes (Molec-

ular Probes, Invitrogen, Carlsbad, CA) or with IRDyes 800CW and

700 DX (LI-COR Biosciences, Lincoln, NE) following the manu-

facturer’s instructions. Labeled proteins were extensively dialyzed

extensively against PBS (pH 7.2) to remove unbound probe and

were more than 99% precipitable with trichloroacetic acid.



Mice were given a single intravenous dose of 2-mg Alexa 594-

albumin and 0.5 mg of Alexa 488-IgG (or respective IRDye Infrared

dyes). Blood and urine samples were collected 3, 12, and 24 h after

injection and daily thereafter. All samples were collected on ice and

stored at �80°F and limited to one freeze-thaw cycle. The quantities

of fluorescence-labeled albumin and IgG were determined spec-

trofluorimetrically (Synergy HT Multi-Mode Microplate Reader,

Biotek, Winooski, VT). Excitation and emission wavelengths were

590 and 617 nm for Alexa 594-albumin, and 494 and 519 nm for Alexa

488-IgG. To minimize variation, sera and urines from a given animal

were measured in the same microtiter plate at the same time. For each

animal, serum and urine obtained before injection were used to de-

termine background fluorescence, which was subtracted from all sub-

sequent values to derive specific fluorescence units per milliliter se-

rum or urine. These were used to calculate logarithmically

transformed clearance curves as well as FE values. For experiments

using IRDye-labeled proteins, urine was collected within 1 h of injec-

tion and separated by SDS-PAGE under reducing and nonreducing

conditions, followed by direct imaging of the gel (Odyssey Infrared

Imaging System, LI-COR Biosciences). Urines collected before injec-

tion were used as controls.

Additional experiments were performed with urines collected for

3 h after injection of unlabeled albumin. These were diluted to 0.3 ml

with PBS and separated over a 1.6 � 60 cm (120 ml) Sephacryl S-100

HR column (GE Healthcare Life Sciences, Amersham). Absorbance at

280 nm was measured continuously and expressed relative to elution

volume (Ve, divided by total column volume, Vt).

Measurements from Serum and UrineSerum and urinary concentrations of albumin and total IgG were

measured by ELISA (Bethyl Laboratories, Montgomery, TX).49,50 Se-

rum and urinary creatinine were measured using a commercially ob-

tained kit according to manufacturer’s instructions (Stanbio Labora-

tory, Boerne, TX). FE of albumin and IgG were calculated from

concentrations of albumin, IgG, and creatinine from serum (S) and

urine (U) samples obtained contemporaneously, using the following

formula:

FEalbumin �([Ualbumin] � [Screatinine])

([Salbumin] � [Ucreatinine])(1)

FEIgG �([UIgG] � [Screatinine])

([SIgG] � [Ucreatinine])(2)

Renal Localization of Albumin and IgGIn some animals, a single intravenous dose of Alexa 594-albumin and

Alexa 488-IgG was given as for clearance studies; these were followed

at varying time intervals by euthanasia and tissue harvesting. Initial

studies were performed in wild-type and FcRn�/� mice to determine

appearances in kidney after intravenous injection of Alexa-labeled

proteins. Tissues were snap-frozen in ice-cold 2-methylbutane. Cry-

ostat sections (4 �m) were fixed in 4% paraformaldehyde in PBS (pH

7.4) and permeabilized with 0.2% Triton X-100. Sections were addi-

tionally stained with Alexa 350-conjugated wheat germ agglutinin

(Molecular Probes, Invitrogen, Carlsbad, CA) to identify glomeruli

and proximal tubules, or goat anti-synaptopodin (Santa Cruz Bio-

technology, Santa Cruz, CA) followed by Alexa 350-labeled donkey

anti-goat IgG (Molecular Probes) to visualize podocytes. All proce-

dures were performed at 4°C with each step taking less than 5 min to

minimize loss of tissue-bound fluorescence probes. Tissues that were

being directly compared were processed contemporaneously. Images

were acquired using an Olympus IX-81 spinning disk confocal micro-

scope.

To determine image intensities within given fields of interest, we

used Image J software (National Institutes of Health, Bethesda, MD)

and a macro essentially as described by the Wright Cell Imaging

Facility (Toronto, ON, Canada; http://www.uhnres.utoronto.ca/

facilities/wcif). The steps used to measure image intensities included

(1) pseudoflat field correction, in which the image was divided by a

blurred version of itself to reveal the illumination pattern; (2) back-

ground correction, using the “rolling ball” algorithm to remove un-

even background from images; (3) setting of a uniform threshold for

all images to distinguish background from true positive staining; and

(4) measurement of an integrated density by summing all pixel inten-

sities above this threshold. To minimize any potential biases, a differ-

ent individual scored the acquired images, which were saved as coded

grayscale files. The first observer was masked to origin of the slides

whereas the second was unaware of the code.

Statistical AnalysisData were analyzed using Minitab statistical software (College Park,

MD). Data sets were first analyzed with the Anderson–Darling nor-

mality test and considered parametric with H� � 0.05. Parametric

data are presented as mean � SEM. Comparisons between nontrans-

planted wild-type and FcRn�/� mice were performed by two-sample

t test, whereas those among transplanted groups were first compared

by one-way ANOVA. If these were significantly different between

groups (P � 0.05), comparisons were made by Fisher’s pairwise test-

ing. Nonparametric data sets were analyzed with the Kruskal–Wallis

Test.

For clearance data, the equation y � 10a0�a1ln(x) was used, in which

x and y were time after injection and fluorescence intensity, respec-

tively. For each animal, a0 and a1 were calculated from log-trans-

formed data using the least-squares method. Curves from individual

animals had r values of 0.91 or more (median 0.987, mean 0.979, n �

33). Individual a0 and a1 values were averaged to generate clearance

curves of albumin and IgG in the different groups.

ACKNOWLEDGMENTS

The authors thank Miglena Petkova and Lauren Hensley for out-

standing technical assistance, and Dr. Vytas Bindokas for assistance

with imaging studies. This work was supported by National Institutes

of Health grants R01DK041873 and R01DK055357 (R.J.Q.).

DISCLOSURESNone.



REFERENCES

1. Simister NE, Mostov KE: An Fc receptor structurally related to MHCclass I antigens. Nature 337: 184–187, 1989

2. Simister NE, Story CM, Chen HL, Hunt JS: An IgG-transporting Fcreceptor expressed in the syncytiotrophoblast of human placenta. EurJ Immunol 26: 1527–1531, 1996

3. Simister NE, Rees AR: Isolation and characterization of an Fc receptorfrom neonatal rat small intestine. Eur J Immunol 15: 733–738, 1985

4. Borvak J, Richardson J, Medesan C, Antohe F, Radu C, Simionescu M,Ghetie V, Ward ES: Functional expression of the MHC class I-relatedreceptor, FcRn, in endothelial cells of mice. Int Immunol 10: 1289–1298, 1998

5. Blumberg RS, Koss T, Story CM, Barisani D, Polischuk J, Lipin A, PabloL, Green R, Simister NE: A major histocompatibility complex classI-related Fc receptor for IgG on rat hepatocytes. J Clin Invest 95:2397–2402, 1995

6. Haymann JP, Levraud JP, Bouet S, Kappes V, Hagege J, Nguyen G, XuY, Rondeau E, Sraer JD: Characterization and localization of the neo-natal Fc receptor in adult human kidney. J Am Soc Nephrol 11:632–639, 2000

7. Jones EA, Waldmann TA: The mechanism of intestinal uptake andtranscellular transport of IgG in the neonatal rat. J Clin Invest 51:2916–2927, 1972

8. Rodewald R: pH-dependent binding of immunoglobulins to intestinalcells of the neonatal rat. J Cell Biol 71: 666–669, 1976

9. Chaudhury C, Brooks CL, Carter DC, Robinson JM, Anderson CL:Albumin binding to FcRn: Distinct from the FcRn-IgG interaction.Biochemistry 45: 4983–4990, 2006

10. Ward ES, Zhou J, Ghetie V, Ober RJ: Evidence to support the cellularmechanism involved in serum IgG homeostasis in humans. Int Immu-nol 15: 187–195, 2003

11. Roopenian DC, Christianson GJ, Sproule TJ, Brown AC, Akilesh S,Jung N, Petkova S, Avanessian L, Choi EY, Shaffer DJ, Eden PA,Anderson CL: The MHC class I-like IgG receptor controls perinatal IgGtransport, IgG homeostasis, and fate of IgG-Fc-coupled drugs. J Im-munol 170: 3528–3533, 2003

12. Chaudhury C, Mehnaz S, Robinson JM, Hayton WL, Pearl DK, Roope-nian DC, Anderson CL: The major histocompatibility complex-relatedFc receptor for IgG (FcRn) binds albumin and prolongs its lifespan. JExp Med 197: 315–322, 2003

13. Tryggvason K, Wartiovaara J: How does the kidney filter plasma?Physiology (Bethesda ) 20: 96–101, 2005

14. Russo LM, Sandoval RM, McKee M, Osicka TM, Collins AB, Brown D,Molitoris BA, Comper WD: The normal kidney filters nephrotic levelsof albumin retrieved by proximal tubule cells: Retrieval is disrupted innephrotic states. Kidney Int 71: 504–513, 2007

15. Kobayashi N, Suzuki Y, Tsuge T, Okumura K, Ra C, Tomino Y: FcRn-mediated transcytosis of immunoglobulin G in human renal proximaltubular epithelial cells. Am J Physiol Renal Physiol 282: F358–F365,2002

16. Akilesh S, Huber TB, Wu H, Wang G, Hartleben B, Kopp JB, Miner JH,Roopenian DC, Unanue ER, Shaw AS: Podocytes use FcRn to clear IgGfrom the glomerular basement membrane. Proc Natl Acad Sci U S A105: 967–972, 2008

17. Hilliard LM, Osicka TM, Clavant SP, Robinson PJ, Nikolic-Paterson DJ,Comper WD: Characterization of the urinary albumin degradationpathway in the isolated perfused rat kidney. J Lab Clin Med 147:36–44, 2006

18. Roopenian DC, Akilesh S: FcRn: The neonatal Fc receptor comes ofage. Nat Rev Immunol 7: 715–725, 2007

19. Israel EJ, Taylor S, Wu Z, Mizoguchi E, Blumberg RS, Bhan A, SimisterNE: Expression of the neonatal Fc receptor, FcRn, on human intestinalepithelial cells. Immunology 92: 69–74, 1997

20. Israel EJ, Wilsker DF, Hayes KC, Schoenfeld D, Simister NE: Increased

clearance of IgG in mice that lack beta 2-microglobulin: Possibleprotective role of FcRn. Immunology 89: 573–578, 1996

21. Ward ES, Martinez C, Vaccaro C, Zhou J, Tang Q, Ober RJ: Fromsorting endosomes to exocytosis: Association of Rab4 and Rab11GTPases with the Fc receptor, FcRn, during recycling. Mol Biol Cell 16:2028–2038, 2005

22. Petkova SB, Akilesh S, Sproule TJ, Christianson GJ, Al KH, Brown AC,Presta LG, Meng YG, Roopenian DC: Enhanced half-life of geneticallyengineered human IgG1 antibodies in a humanized FcRn mousemodel: Potential application in humorally mediated autoimmune dis-ease. Int Immunol 18: 1759–1769, 2006

23. Kim J, Bronson CL, Hayton WL, Radmacher MD, Roopenian DC,Robinson JM, Anderson CL: Albumin turnover: FcRn-mediated recy-cling saves as much albumin from degradation as the liver produces.Am J Physiol Gastrointest Liver Physiol 290: G352–G360, 2006

24. Bertolatus JA, Hunsicker LG: Glomerular sieving of anionic and neutralbovine albumins in proteinuric rats. Kidney Int 28: 467–476, 1985

25. Tay M, Comper WD, Singh AK: Charge selectivity in kidney ultrafiltra-tion is associated with glomerular uptake of transport probes. Am JPhysiol 260: F549–F554, 1991

26. Yu W, Sandoval RM, Molitoris BA: Rapid determination of renal filtra-tion function using an optical ratiometric imaging approach. Am JPhysiol Renal Physiol 292: F1873–F1880, 2007

27. Schaeffer RC Jr, Gratrix ML, Mucha DR, Carbajal JM: The rat glomer-ular filtration barrier does not show negative charge selectivity. Mi-crocirculation 9: 329–342, 2002

28. Birn H, Christensen EI: Renal albumin absorption in physiology andpathology. Kidney Int 69: 440–449, 2006

29. Eppel GA, Osicka TM, Pratt LM, Jablonski P, Howden B, Glasgow EF,Comper WD: The return of glomerular filtered albumin to the rat renalvein: The albumin retrieval pathway. Ren Fail 23: 347–363, 2001

30. Koltun M, Comper WD: Retention of albumin in the circulation isgoverned by saturable renal cell-mediated processes. Microcirculation11: 351–360, 2004

31. Koltun M, Nikolovski J, Strong K, Nikolic-Paterson D, Comper WD:Mechanism of hypoalbuminemia in rodents. Am J Physiol Heart CircPhysiol 288: H1604–H1610, 2005

32. Osicka TM, Strong KJ, Nikolic-Paterson DJ, Atkins RC, Jerums G,Comper WD: Renal processing of serum proteins in an albumin-deficient environment: An in vivo study of glomerulonephritis in theNagase analbuminaemic rat. Nephrol Dial Transplant 19: 320–328,2004

33. Greive KA, Nikolic-Paterson DJ, Guimaraes MA, Nikolovski J, PrattLM, Mu W, Atkins RC, Comper WD: Glomerular permselectivity factorsare not responsible for the increase in fractional clearance of albuminin rat glomerulonephritis. Am J Pathol 159: 1159–1170, 2001

34. Russo LM, Bakris GL, Comper WD: Renal handling of albumin: Acritical review of basic concepts and perspective. Am J Kidney Dis 39:899–919, 2002

35. Mancilla-Jimenez R, Appay MD, Bellon B, Kuhn J, Bariety J, Druet P:IgG Fc membrane receptor on normal human glomerular visceralepithelial cells. Virchows Arch A Pathol Anat Histopathol 404: 139–158, 1984

36. Michael AF Jr, Drummond KN, Good RA, Vernier RL: Acute poststrep-tococcal glomerulonephritis: Immune deposit disease. J Clin Invest45: 237–248, 1966

37. Couser WG, Abrass CK: Pathogenesis of membranous nephropathy.Annu Rev Med 39: 517–530, 1988

38. Kazatchkine MD, Fearon DT, Appay MD, Mandet C, Bariety J: Immu-nohistochemical study of the human glomerular C3b receptor in nor-mal kidney and in seventy-five cases of renal diseases. J Clin Invest 69:900–912, 1982

39. Alexander JJ, Wang Y, Chang A, Jacob A, Minto AW, Karmegam M,Haas M, Quigg RJ: Mouse podocyte complement factor H: The func-tional analogue to human complement receptor 1. J Am Soc Nephrol18: 1157–1166, 2007



40. Reiser J, von GG, Loos M, Oh J, Asanuma K, Giardino L, Rastaldi MP,Calvaresi N, Watanabe H, Schwarz K, Faul C, Kretzler M, Davidson A,Sugimoto H, Kalluri R, Sharpe AH, Kreidberg JA, Mundel P: Inductionof B7-1 in podocytes is associated with nephrotic syndrome. J ClinInvest 113: 1390–1397, 2004

41. Schiffer M, Mundel P, Shaw AS, Bottinger EP: A novel role for theadaptor molecule CD2-associated protein in transforming growth fac-tor-beta-induced apoptosis. J Biol Chem 279: 37,004–37,012, 2004

42. Mendrick DL, Kelly DM, Rennke HG: Antigen processing and presen-tation by glomerular visceral epithelium in vitro. Kidney Int 39: 71–78,1991

43. Zhu X, Meng G, Dickinson BL, Li X, Mizoguchi E, Miao L, Wang Y,Robert C, Wu B, Smith PD, Lencer WI, Blumberg RS: MHC classI-related neonatal Fc receptor for IgG is functionally expressed inmonocytes, intestinal macrophages, and dendritic cells. J Immunol166: 3266–3276, 2001

44. Bergtold A, Desai DD, Gavhane A, Clynes R: Cell surface recycling ofinternalized antigen permits dendritic cell priming of B cells. Immunity23: 503–514, 2005

45. Soos TJ, Sims TN, Barisoni L, Lin K, Littman DR, Dustin ML, Nelson PJ:

CX3CR1� interstitial dendritic cells form a contiguous networkthroughout the entire kidney. Kidney Int 70: 591–596, 2006

46. Cunningham PN, Dyanov HM, Park P, Wang J, Newell KA, Quigg RJ:Acute renal failure in endotoxemia is caused by TNF acting directly onTNF receptor-1 in kidney. J Immunol 168: 5817–5823, 2002

47. Cunningham PN, Wang Y, Guo R, He G, Quigg RJ: Role of Toll-likereceptor 4 in endotoxin-induced acute renal failure. J Immunol 172:2629–2635, 2004

48. Bao L, Wang Y, Chang A, Minto AW, Zhou J, Kang H, Haas M, QuiggRJ: Unrestricted C3 activation occurs in Crry-deficient kidneys andrapidly leads to chronic renal failure. J Am Soc Nephrol 18: 811–822,2007

49. Bao L, Haas M, Boackle SA, Kraus DM, Cunningham PN, Park P,Alexander JJ, Anderson RA, Culhane C, Holers VM, Quigg RJ: Trans-genic expression of a soluble complement inhibitor protects againstrenal disease and promotes survival in MRL/lpr mice. J Immunol 168:3601–3607, 2002

50. Quigg RJ, Lim A, Haas M, Alexander JJ, He C, Carroll MC: Immunecomplex glomerulonephritis in C4- and C3-deficient mice. Kidney Int53: 320–330, 1998



Renal FcRn Reclaims Albumin but Facilitates Elimination of IgG

Documents

Transcript of Renal FcRn Reclaims Albumin but Facilitates Elimination of IgG