Kidney International, Vol. 37 (1990), pp. 1395—1401

EDITORIAL REVIEW

Tamm-Horsfall protein—uromodulin (1950—1990)

Tamm-Horsfall glycoprotein (THP) is the most abundantprotein in normal human urine. It was first described as"urinary mucoprotein" by Morner [1] and further characterizedwhen Tamm and Horsfall [21 described the salt precipitationmethod of isolating THP from urine and found it to be thesubstance responsible for urinary inhibition of myxovirus-induced hemagglutination. It is present in the kidneys of allplacental mammals but has not been demonstrated in marsupi-als, monotremes, or other vertebrates [3]. THP is a majorconstituent of urinary casts [4—6], and in humans, the dailyexcretion ranges from 20 to 200 mg [7]. Its physicochemical andbiological properties have been studied extensively, but itsfunction has remained obscure [6, 81.

In 1985, another glycoprotein was independently purifiedfrom urine of pregnant women using lectin adherence columnsand called uromodulin [9]. The carbohydrate moiety of uromod-ulin subsumes numerous functions. It is a specific ligand forcytokines, such as interleukin-l (IL-l), interleukin-2 (IL-2), andtumor necrosis factor (TNF). Related oligosaccharides inhibitin vitro T cell proliferation induced by specific antigens. Otherattached carbohydrates are responsible for inhibition of anIL-i-dependent co-mitogenic assay. Recently, we and othersdemonstrated by cDNA and amino acid sequencing that theprotein backbone of THP and uromodulin are identical [10, 11].To emphasize the importance of glycosylation, we refer to theglycoprotein isolated by the salt precipitation method of Tammand Horsfall as THP and that isolated from pregnancy urine bylectin adherence as uromodulin [9].

BiochemistryUrinary THP exists in a polymeric form (Mr 7 x i0 D),

which can be dissociated into monomeric molecules (Mr 80,000D) with urea [12], 50% acetic acid [13], 6 M guanidine hydro-chloride, and 0.05% sodium dodecyl sulfate [61. The amino acidcomposition was estimated by Fletcher et al [14] and revealedno unusual amino acids. Acidic amino acids outnumbered thebasic amino acids, giving THP a low isoelectric point.

Recent cloning and sequencing of the THP (uromodulin) genehas essentially confirmed the amino acid composition reportedpreviously [10]. Uromodulin clones were isolated from a humankidney cDNA library using oligonucleotide probes that weresynthesized based on extensive amino acid sequence dataoriginating from tryptic digests of THP and uromodulin. Avail-able evidence suggests the presence of a single THP gene.Nucleotide sequencing of a full-length cDNA predicts a proteinof 640 amino acids. This precursor protein includes a 24 amino

Received for publication October 5, 1988and in revised form January 22, 1990Accepted for publication January 23, 1990

© 1990 by the International Society of Nephrology

acid leader sequence, which is cleaved during processing,resulting in a mature protein with 616 amino acids including 48cysteine residues. Hydrophobicity plotting shows that THPdoes not have a classic transmembrane hydrophobic region.This, along with the presence of a characteristic leader se-quence, suggests that the majority of THP is in the form of asecreted protein. However, this finding also raises an interest-ing dilemma: evidence based on microscopic studies, ultra-structural studies, and antibody cross-linking and capping phe-nomena all strongly suggest that THP exists, at least in part, asa membrane-associated protein. This makes the membraneform of THP an excellent candidate to be anchored to lipid viaa phospholipase C-sensitive linkage as previously described foralkaline phosphatase, 5' nucleotidase, decay accelerating fac-tor, scrapie prion protein, the LFA-3 antigen, and the Thylantigen [15]. These studies for THP remain to be done. Fourseparate homologous domains of THP exhibit strong similarityto the cysteine-rich region of the epidermal growth factorprecursor [16]. This domain has been described in a number ofextracellular glycoproteins, including the LDL receptor, throm-bomodulin, tissue plasminogen activator, and a number ofclotting factors, including Factor IX, Factor X, and protein S[17]. Recent studies suggest that a subset of these cysteine-richdomains acts as a consensus sequence leading to the post-translational hydroxylation of aspartic acid and asparagine [18].Indeed, both potential asparagines in the THP molecule arehydroxylated. THP (uromodulin) also expresses an "RGD"sequence previously described in a number of extracellularmatrix proteins, such as fibronectin, fibrinogen, type 1 collagen,and thrombospondin, which bind to cell surface receptors of theintegrin superfamily [19, 20]. Thus, the THP gene containsseveral conserved structural and functional domains commonto many important groups of proteins.

The carbohydrate moiety accounts for 30% of the THPmolecule by weight [4, 9]. Based on the cDNA predicted aminoacid sequence analysis, human THP has eight potential N-linked glycosylation sites. However, analytical data suggestthat only five are utilized [21]. There is no evidence for 0-linkedsugars in human THP [22]. The carbohydrate moiety is essentialfor inhibition of viral hemagglutination, the property that led tothe initial characterization of the protein by Tamm and Horsfall[2]. This property probably resides in the nonacetylated sialicacid component, because THP produced by hamster kidneycells, which lacks sialic acid, and rabbit THP, which was0-acetylated sialic acid, do not inhibit viral hemagglutination[23, 24].

THP in solution tends to gel readily [25]; this property hasobvious implications for urinary cast formation, and intactglycosylation appears to be important for this property. Gelformation is increased by various factors, such as calcium andsodium ions [25], albumin [26], radiocontrast media [27—29],

1396 Kumar and Muchmore: Tamm-Horsfall protein—uromodulin

Sd antigen /'GaINac.-GaI—- GIcNac/

NANA (2-3)

Dolichos biflorusbinding site

Key

[: :;i::::

and Bence Jones proteins [30]. It is reduced by urea [12] and byalkaline pH [25].

THP from 90% of the Caucasian population expresses aspecific structural complex consisting of a 13-linked N-acety-lated galactosamine; this substance, also found on red bloodcells, is known as Sd antigen and is the determinant of the Sdblood group [31, 32]. This antigen may explain the reportedimmunological cross reactivity between THP and erythrocyteghosts [33].

Approximately 90% of the oligosaccharides expressed onhuman THP consist of complex sialylated tn- and tetra-anten-nary chains [21]. THP also expresses the specific binding sitefor L-PHA as well as polylactosamine termini measured bothstructurally and by the affinity of THP for the lectin Dolichosbiflorus [34] (Fig. lA). Evidence by Serafini-Cessi, Dall'olio andMalagolini [35] supports the earlier suggestion [22] that THPalso expresses unprocessed mannose-rich chains of the generalstructure GlcNac2Man(57). High-resolution NMR data confirmthe presence of M5, M6, and M7 oligosaccharides (Fig. 1) [36].Uromodulin, compared to THP, is rich in these unprocessedhigh mannose chains. Our evidence suggests that MS,M6 car-bohydrate residues are critical for binding to certain cytokines,

A

Fig. 1. A. Structure of idealized tetra-antennary chain of THPdemonstrating multiple lectin binding sites. Evidence [21] suggestssignificant heterogeneity in these chains. B. Structure of mannose-richchains on THP. M5, M6, and M7 structures are based on H-nmr data [36].D-mannose, Man or (M); D galactose, Gal; N-acetyl glucosamine, GlcNac;N-acetyl galactosamine, GalNac; sialic acid, NANA; asparagine, asn;linkage key as noted in Figure.

B M5

M-M_+

M GIcNac GIcNac AsnM

,1M

M6

M

M

1M GIcNac GIcNac - Asn

M

M

4M

Mi

M+M

MM GIcNac GIcNac Asn

>1M

+M

4M

including IL-l, IL-2, and TNF. Furthermore, these same gly-cosylation modifications are able to mimic the in vitro immu-nomodulatory function of uromodulin [37]. Functional evidencesuggests that alterations in the carbohydrate structure mayoccur in pregnancy [10]. Uromodulin, which was originallyisolated by lectin (concanavalin A) adherence columns fromhuman pregnancy urine based on its in vitro immunosuppres-sive activity, is at least 10 times more immunosuppressive invitro than salt-precipitated THP. It appears that differentialbioactivity is not due to amino acid sequence differences, but islikely the result of differences in glycosylation between THP(isolated from nonpregnancy urine by salt precipitation) anduromodulin (isolated from pregnancy urine using a lectin adher-ence column) [10]. Alteration of THP glycosylation has alsobeen described in diabetes [38].

Localization

THP distribution in the body is primarily limited to thekidney. Using Northern blot analysis of total RNA from variousrat tissues, significant message was detected only in extracts ofkidney; liver, heart, lung, brain, thymus, muscle, spleen andtestis were negative. Human placenta mRNA also lacked mes-

Kumar and Muchmore: Tamm-Horsfall protein—uromodulin 1397

eight weeks in human fetal kidney, two days pre-term in rat[47], and three days pre-term in hamster kidney [48]. Urinefrom fetal kidney contributes to amniotic fluid from 12 to 14weeks [49]. While THP has been described in human amnioticfluid near term by several researchers, studies in mid-termamniotic fluid have been more equivocal [50—52]. We testedamniotic fluid from four patients at 16 weeks and did not findTHP by an enzyme-linked immunosorbent assay in which thelower limit of detection was 20 ng/ml (S. Kumar and P.Schmidt, unpublished observation).

The relative abundance, specific nephronal location, andevolutionary conservation of THP suggest that this glycopro-tein has important physiologic functions. Until recently, thespeculation regarding the possible function of THP has re-volved around the physiological properties of TAL. It has beensuggested that the remarkable gel-forming tendency of THPmight be responsible for the water impermeability of TAL, withthe absence of THP from macula densa serving as a windowallowing tubuloglomerular feedback to occur. Furthermore, thepresence of RGD sequence in the THP molecule makes itpossible that THP acts as an intercellular adhesion protein bybinding to integrin receptors on adjacent cell surfaces. Thisalluring hypothesis remains to be tested experimentally.

Although early reports suggested that THP binds to furo-semide [53], ethacrynic acid, and mercuric chloride [8], morerecent studies have not confirmed this finding [54]. Since thesediuretic agents act on TAL, it had been postulated that THPmay be the 2C1-K-Na co-transporter, but there is no directevidence for this speculation. An immunochemical study of avariety of rat epithelia found that antiserum to THP did notco-localize with known intracellular distribution of the 2C1-K-Na co-transporter [55]. The uniform presence of THP onboth luminal and basolateral aspects of the TAL cells arguesagainst it being a protein involved in vectorial transport of ionsacross the epithelium.

Better evidence is available to support a role for THP inurothelial defense against infection, a concept first suggested byOrskov, Ferencz and Orskov [56, 57] when they showed that E.coli with type I fimbriae were trapped by THP. These findingshave been confirmed by other groups [58—60]. Kuriyama andSilverblatt [59] also demonstrated that bacteria coated by THPare less susceptible to phagocytosis by polymorphonuclearlymphocytes compared with uncoated bacteria. This binding toE. co/i type I pili is dependent on the mannose content of THP,and exogenous mannose can successfully compete for it in invitro binding assays [56, 57].

Recent studies have markedly expanded the potential immu-nomodulatory role of THP and suggest that this glycoproteinmay play a central role in the regulation of circulating levels andbiologic activity of several important cytokines, including IL-l,TNF, and IL-2 [61, 62]. The original studies suggesting animportant role for this glycoprotein were performed with uro-modulin. Uromodulin was first shown to be a high affinity ligandfor rIL-l [63], and later work demonstrated that rTNF and rIL-2also bind specifically to uromodulin [62, 64]. Subsequent workhas shown that THP isolated by salt precipitation shares anidentical binding affinity with uromodulin. All of these studies,however, have utilized immobilized cytokines as a capture

Afferent arteriole

Proximalconvoluted

tubulePhysiologic function

llectingduct

Thin limbsof loop of

Henle

Fig. 2. Distribution of THP in the human nephron is limited to the thickascending limb and the early distal convoluted tubule. THP is absentfrom the macula densa.

sage [10]. Within the kidney, THP is specifically localized (Fig.2) to the epithelial cells of the thick ascending limbs (TAL) ofthe loops of Henle and the most proximal part of the distalconvoluted tubule, which is thought to be embryologically andfunctionally similar to the TAL [39, 40]. This protein is char-acteristically absent from macula densa and from glomeruli,proximal convoluted tubules, thin limbs of the loops of Henle,collecting duct, blood vessels, and the interstitium [41, 42]. Onimmunoelectron microscopy, THP is found in both luminal andbasolateral aspects of the cell membrane and in the membranesof Golgi apparatus and endoplasmic reticulum [8, 41]. Smallquantities of THP can be detected in blood by sensitive immu-noassays [8, 43, 44]. The finding of THP immuno-cross-reactivematerial from liver and cerebrospinal fluid using antisera toTHP has been interpreted to suggest that THP might be presentextrarenally in all tissues involved in active chloride transport[45], but these studies have not been confirmed using monoclo-nal antibodies [42], nor has mRNA for THP been detected inthese tissues [10]. However, the renal environment appears notto be essential for transcription of the THP gene and forsynthesis of this protein. Pieces of rat renal cortex and medulla,when transplanted to the anterior chamber of the eye, prolifer-ate to form tubule-like structures [46]. Sections of these im-plants reacted with monoclonal antibodies specific for rat THPusing an immunoperoxidase technique (S. Kumar and M. Celio,unpublished observation).

Developmentally, the appearance of THP coincides withmaturation of the thick ascending limb of Henle's loop. Immu-nohistochemical studies have demonstrated its presence at

1398 Kumar aud Muchmore: Tamm-Horsfall protein—uromodulin

reagent. Moonen and colleagues attempted to reproduce thesestudies with both uromodulin and TNF in solution. Theseauthors found that rTNF and uromodulin in solution only boundto each other at a pH below 6.0 with optimal interactionoccurring at pH 5.0, conditions likely to be found in the distaltubule [65]. Furthermore, these authors suggested that TNFmight be inactive under these conditions, but more recentstudies suggest that TNF might actually have enhanced activityat a low pH [66]. In all cases, these cytokines are binding tospecific carbohydrate sequences found on THP. This conclu-sion is based on four different experimental approaches. First,oxidative destruction of exposed carbohydrates on THP orenzymatic removal with N-glycanase destroys the ability ofTHP to bind to cytokines [63]. Second, defined mono- anddisaccharides specifically compete with THP for binding toimmobilized IL-l, TNF, and IL-2 [62, 64]. Third, defined plantlectins, such as concanavalin A and wheat germ agglutinin,compete with IL-i and TNF for binding to the carbohydratemoiety of THP [62]. Finally, using a panel of IL-i /3 analoguesengineered by site-specific mutagenesis and differing from oneanother by a single amino acid substitution, we have developedevidence that IL-l f3 has two different binding domains (A.Muchmore and A. Shaw, unpublished observation). One bind-ing domain is lectin-like and binds specifically to carbohydrateexpressed by THP; the alterations of this site affect binding toTHP but do not affect binding of IL- 1 to the 80-kD, cell surfaceIL-l receptor that was first described by Dower et al [67].Conversely, substitutions at the cell surface receptor bindingsite affect cell binding but do not alter binding to THP [67, 68].

Immunohistochemical studies demonstrate that exogenousrIL-l and rTNF bind to renal tubular segments that expressTHP, namely TAL. Mutant IL-i that fails to bind to thecarbohydrate domain of THP in vitro also fails to bind to therenal tubule sections (A. Muchmore, unpublished observation).Most studies that have examined the in vivo administration of'25l-labeled rIL-l, rTNF, or rIL-2 demonstrate that the kidneyis the major organ responsible for the rapid (Tl/2 on average of5 to 10 minutes) plasma clearance of these cytokines (A.Muchmore, unpublished observation) [69—72]. However, somestudies show more widespread tissue distribution, with thekidney being only one of several organs responsible for uptake[73]. In aggregate these observations suggest that THP ex-presses carbohydrate moieties recognized by a number ofcytokines and that the interaction between THP and thesecytokines is optimal at an acidic pH. Since the kidney appearsto be a major site of cytokine catabolism, these studies offer apotential mechanism which deserves further study.

The immunoregulatory role of THP has been somewhatcontroversial. Early results suggested that THP induced T cellproliferation [74], while later studies found that THP inhibitedlectin-induced proliferation and the mixed lymphocyte response[75]. These authors found that the immunoregulatory activity ofTHP resides in the carbohydrate moiety. Our own studies usinguromodulin demonstrate that it inhibits T cell proliferation inassays of IL-i-induced proliferation and antigen-specific prolif-eration [10]. However, neither THP nor uromodulin blocks theability of IL- 1 to act directly on fibroblasts to induce prosta-glandin synthesis (A. Muchmore, unpublished observation).Furthermore, uromodulin fails to block binding of IL-i to itscell surface receptor (K. Matsushima and J. Oppenheim, per-

sonal communication). This is consistent with the observationthat the uromodulin binding site on rIL-l /3 is physically distinctfrom the EL-4 murine binding site of rIL-l. Studies using anIL-l co-mitogenic assay demonstrate that excess phytohemag-glutinin (PHA), but not excess IL-l, will overcome inhibitionmediated by either uromodulin or THP. Thus, although immo-bilized IL-I binds to uromodulin with a very high affinity,inhibition in the IL-l co-mitogenic assay is the result of com-petition with PHA and not with IL-l. These results confirm thehypothesis of Serafini-Cessi et al [76], who suggested that THPmight compete for a cell surface lectin (PHA) binding site.Interestingly, oligosaccharides that inhibit the IL-l/PHA co-mitogenic assay fail to inhibit T cell proliferation induced byspecific antigen and vice versa. We have purified the oligosac-charides derived from pregnancy THP (uromodulin) responsi-ble for inhibition of antigen-specific proliferation and structur-ally characterized them using high resolution nuclear magneticresonance (A. Sherblom, H. van Halbeek, and A. Muchmore,unpublished data). These homogenous high-mannose oligosac-charides inhibit antigen-specific proliferation at concentrationsof approximately 50 LM. Closely related oligosaccharides di-rectly compete with uromodulin for binding to rIL-l, rTNF, andrIL-2 at concentrations as low as 500 nM. PHA-binding oh-gosaccharides (responsible for inhibition of the IL-i co-mito-genic assay) fail to inhibit either the antigen-specific T cellproliferative assay or the lymphokine binding assay. These datasuggest that oligosaccharides isolated from uromodulin havemultiple functions.

Significance of THP in disease states

Interstitial deposits of THP have been found in obstructiveuropathy [77], rejecting renal transplants [78, 79] and in neph-ronothiasis [80, 81]. These deposits are thought to occurthrough leakage of THP from rupture of obstructed or damagedtubules. However, these deposits do not show consistent cor-relation with inflammatory reaction and their significance inpathogenesis of tubulo-interstitial scarring is doubtful [82].Deposits of THP have also been demonstrated in the glomerularmesangium in experimental models of vesico-ureteric reflux inrats, suggesting a possible involvement of THP in glomerularscarring associated with reflux nephropathy [83], but moredirect evidence for this hypothesis is lacking.

Injection of THP in rats leads to an inflammatory reactionaround the TAL [841. There is evidence that this reaction iselicited by in situ formation of immune complexes. In animalspassively immunized with anti-THP antibodies, these com-plexes are cleared more quickly [85]. Similar rabbit models ofTHP-induced tubulo-interstitial nephritis have also been devel-oped [86, 87], but there is no known human disease counterpartto these experimental conditions.

THP has been conclusively shown to be present in the core ofrenal stones [88]. However, it is not clear whether it is aninnocent bystander or an active participant in stone formation.In vitro assays of calcium stone growth have produced variedresults. Some have shown THP to inhibit stone growth [89—93],while others have shown promotion [94, 95], and yet othershave shown no effect [96]. We studied this effect in a "Ccalcium oxalate seed crystal growth assay and found that THPhad no effect on calcium oxalate crystal growth [97]. Interest-ingly, Keutel [98], who had provided some of the early evidence

Kumar and Muchmore: Tamm-Horsfall protein—uromodulin 1399

for the presence of THP in stones, also reported failure todetect THP in the urine of African Blacks and related thisfinding to the reported low incidence of renal stones in thispopulation [99]. We tested the urine of five Black Africans byELISA employing a monoclonal antibody to THP and detectednormal quantities of THP (S. Kumar, unpublished observation).

THP is a major constituent of urinary casts. Conditionspromoting polymerization of THP stimulate cast formation.Experimental and clinical evidence suggests that tubular ob-struction by urinary casts contributes to development andmaintenance of acute renal failure [reviewed in 100, 101]. Invitro studies have demonstrated increased aggregability of THPwith Bence Jones protein and radiocontrast media [27—30]. Incast nephropathy associated with multiple myeloma, THP hasbeen demonstrated in Bowman's space, suggesting reflux ofurine in a nephron obstructed distally by a cast [102]. In infantsand children with prolonged nephrograms after intravenousurography, large amounts of THP were found in urine duringsubsequent diuresis. In one infant who died of anuric renalfailure after a dye study, autopsy showed proteinaceous castsplugging renal tubules [103]. THP polymerization in the tubulesmay well be the pathophysiological basis of acute renal failureseen after radiocontrast dye studies in patients with multiplemyeloma and/or volume depletion [104]. Maintenance of highurine volume and urinary alkalization should theoretically pre-vent intratubular cast formation and may be the reason whyfluid therapy has been found to be beneficial in acute renalfailure associated with use of radiocontrast media [1051 andmethotrexate [106], especially since the pharmacokinetics ofmethotrexate itself have been shown not to be altered either byhydration [1071 or by alkalization [106].

Conclusions

Although its specific nephronal location and gel-formingcharacteristics have long suggested a role for THP in renal saltand water transport physiology, this hypothesis remains uncon-firmed. Our knowledge of the molecular biology of THP hasrecently undergone rapid advancement. The complete aminoacid sequence deduced from a cDNA sequence is now avail-able. THP (uromodulin) isolated from pregnant women is apotent immunoregulatory molecule. THP from both pregnantand nonpregnant sources has been shown to be a specific ligandfor a number of potent cytokines, including Il-I, TNF, and IL-2.Both the binding activity to cytokines and the biologic activityof THP are critically dependent upon unique glycosylationpatterns. It is possible that THP could play an important role inthe regulation of circulating levels and potentially the intrarenalbioactivity of many important cytokines. In conclusion, it isinteresting to note that the role of THP carbohydrate and itsability to interact with lectin-like molecules responsible formyxovirus-induced hemagglutination were used in 1950 to firstpurify this glycoprotein. It would appear that, 40 years later,research has come full circle and again stresses the importanceof post-translational changes for this most interesting glycopro-tein.

SATISH KUMAR and ANDREW MUCHMOREChicago, Illinois and Bethesda, Maryland, USA

Reprint requests to Dr. A. Muchmore, Bldg. /0, Rm. 6B09, NIH,Bethesda, Maryland 20892, USA.

References

1. MORNER KAH: Untersuchungen uber die Proteinstoffe und dieeiweissfallenden substanzen des normalen Meschenharns. SkandArch Physiol 6:332—337, 1895

2. TAMM I, HORSFALL FL: Characterisation and separation of aninhibitor of viral hemagglutination present in urine. Proc Soc ExpBiol Med 74:108—114, 1950

3. WALLACE AC, NAIRN RC: Tamm Horsfall protein in kidneys ofhuman embryos and foreign species. Pathology 3:303—310, 1971

4. FLETCHER AP, NEUBERGER A, RATCLLFFE WA: Tamm Horsfallurinary glycoprotein: The chemical composition. Biochem J 120:417—424, 1970

5. FLETCHER AP, NEUBERGER A, RATCLIFFE WA: Tamm Horsfallurinary glycoprotein: The subunit structure. Biochem J 120:425—432, 1970

6. HOYER JR, SELLER MW: Pathophysiology of Tamm-Horsfall pro-tein. Kidney mt 16:279—289, 1979

7. HUNT iS, MCGIVEN AR, GROUFSKY A, LYNN KL, TAYLOR MC:Affinity-purified antibodies of defined specificity for use in asolid-phase microplate radioimmunoassay of human Tamm-Hors-fall glycoprotein in urine. Biochem J 227:957—963, 1985

8. RoNco P. BRuNI5H0Lz M, GENITEAU-LEGENDRE M, CI-IATELETF, VERROUST P, RICHET G: Pathophysiologic aspects of Tamm-Horsfall protein: A phylogenetically conserved marker of the thickascending limb of Henle's loop. Adv Nephrol 16:231—250, 1987

9. MUCHMORE A, DECKER JM: Uromodulin: A unique 85 kilodaltonimmunosuppressive glycoprotein isolated from urine of pregnantwomen. Science 229:479—481, 1985

10. HESSLON C, DECKER JM, SHERBLOM A, KUMAR S, YUE C,MATTALIANO R, TIZARD R, KAWASI-LIMA E, SCHMEISSNER U,HELETKY S, CHOW P. BURNE C, SHAW A, MUCHMORE A:Uromodulin (Tamm Horsfall glycoprotein): A renal ligand forlymphokines. Science 237:1479—1484, 1987

11. PENNICA D, KOHR WJ, KUANG W-J, GLAISTER D, AGGARWALBB, CHEN EY, GOEDDEL DV: Identification of human urornodulinas the Tamm Horsfall urinary glycoprotein. Science 236:83—88,1987

12. CURTAIN CC: The action of urea on the urinary inhibitor ofinfluenza virus hemagglutination. Aust J Exp Biol 31:615—622, 1953

13. FRIEDMANN T, JOHNSON P: The disaggregation of Tamm Horsfallmucoprotein by acetic acid. Biochim Biophys Acta 121:292—297,1966

14. FLETCHER AP, MCLAUGHLIN JE, RATCLIFFE WA, WOODS PA:The chemical composition and electron microscopic appearanceof a protein derived from urinary casts. Biochim Biophys Ada214:299—308, 1970

15. Low MG, SALTIEL AR: Structural and functional roles of glyco-sylphosphatidylinositol in membranes. Science 239:268—275, 1988

16. SUDHOF TC, RUSSELL DW, GOLDSTEIN J, BROWN MS, SANCHEZ-PESCADOR R, BELL GI:Cassette of eight exons shared by genes forLDL receptor and EGF precursor. Science 228:893—895, 1985

17. SUDI-LOF TC, GOLDSTEIN JL, BROWN MS, RUSSELL DW: TheLDL receptor gene: A mosaic of exons shared with differentproteins. Science 228:815—822, 1985

18. PRZYSIECKI CT, STAGGERS JE, RAMJIT HG, MussoN DG, STERNAM, BENNETT CD, FRIEDMAN PA: Occurrence of/3-hydroxylatedasparagine residues in non-vitamin K-dependent proteins contain-ing epidermal growth factor-like domains. Proc Nail Acad SdUSA 84:7856—7860, 1987

19. HYNES R: Integrins: A family of cell surface receptors. Cell48:549—554, 1987

20. RUOSLAHTI E, PIERSCHBACHER MD: New perspective in celladhesion: RGD and integrins. Science 238:491—497, 1987

21. WILLIAMS J, MARSHALL R, VAN HALBEEK H, VLIEGENTHARTJFG: Structural analysis of the carbohydrate moieties of humanTamm Horsfall glycoprotein. Carbohydr Res 134:141—155, 1984

22. ALFONSO A, CHARLWOOD P, MARSHALL RD: Isolation and char-acterization of glycopeptides from digests of human Tamm Hors-fall glycoprotein. Carbohydr Res 89:309—319, 1981

1400 Kumar and Muchmore: Tamm-Horsfall protein—uromodulin

23. NEUBERGER A, RATCLLFFE WA: The acid and enzymic hydrolysisof 0-acetylated sialic acid residues from rabbit Tamm Horsfallglycoprotein. Biochem J 129:683—693, 1972

24. BLOOMFIELD FJ, DUNSTAN DR, FOSTER CL, SEaFINI-CEssI F,MARSHALL RD: Some factors affecting the production by culturedbaby-hamster kidney cells of BHK glycoprotein I which cross-reacts immunologically with Tamm-Horsfall glycoprotein. Bio-chemf 164:41—51, 1977

25. STEVENSON FK, CLEAVE AJ, KENT PW: The effect of ions on theviscometric and ultracentrifugal behaviour of Tamm Horsfallglycoprotein. Biochim Biophys Acta 236:59—66, 1971

26. PESCE AJ, KANT KS, CLYNE DH, POLLAK VE: A model ofurinary cast formation. (abstract) C/in Chem 23:1146, 1977

27. SCHWARTZ RH, BERDON WE, WAGNER J, BECKER J, BAKER DH:Tamm-Horsfall urinary mucoprotein precipitation by urographiccontrast agents: In vitro studies. Am J Roentgenol Radium TherNucI Med 108:698—701, 1970

28. DAWSON P, FREEDMAN DB, HOWELL MJ, HINE AL: Contrast-medium-induced acute renal failure and Tamm-Horsfall protein-uria. Br J Radiol 57:577—579, 1984

29. DAWNAY AB, THORNLEY C, NOCKLER I, WEBB JA, CATTELLWR: Tamm-Horsfall glycoprotein excretion and aggregation dur-ing intravenous urography. Relevance to acute renal failure.Invest Radio! 20:53—57, 1985

30. CLYNE DH, KANT KS, PESCE AJ, POLLAK V: Nephrotoxicity oflow molecular weight serum proteins: Physiochemical interactionsbetween myoglobin, hemoglobin, Bence-Jones proteins andTamm-Horsfall mucoprotein. Curr Probi C/in Biochem 9:299—308,1979

31. DONALD AS, YATES AD, SOH CP, MORGAN W, WATKINS WM: Ablood group SDA active pentasaccharide isolated from TammHorsfall urinary glycoprotein. Biochem Biophys Res Commun115:625—631, 1983

32. DONALD AS, FEENEY J: Oligosaccharides obtained from a bloodgroup Sd(a) Tamm Horsfall glycoprotein. An n.m.r. study.Biochm J 236:821—828, 1986

33. HARTMANN L, DELAUNAY J, OLLIER-HARTMANN MP, BRIN-GUIER A, RICHET G: Tamm-Horsfall protein and erythrocyteghosts immunologically cross react. Biomedicine 35:1—3, 1981

34. So CPC, MORGAN WTJ, WATKINS WM, DONALD ASR: Therelationship between the N-acetyl galactosamine content and theblood group Sd activity of Tamm and Horsfall urinary glycopro-tein. Biochem Biophys Res Commun 93:1132—1139, 1980

35. SERAFINI-CESSI F, DALL'OLIO F, MALAGOLINI N: High-mannoseoligosaccharides from human Tamm Horsfall glycoprotein. BiosciRep 4:269—274, 1983

36. DALL'OLIO F, DE KANTER FJJ, VAN DEN EIJNDEN DH, SER-AFINI-CESSI F: Structural analysis of the preponderant high-mannose oligosaccharide of human Tamm-Horsfall glycoprotein.CarbohydrRes 178:327—332, 1988

37. MUCHMORE AV, SHIFRIN 5, DECKER JM: In vitro evidence thatcarbohydrate moieties derived from uromodulin, an 85,000-daltonimmunosuppressive glycoprotein isolated from human pregnancyurine, are immunosuppressive in the absence of intact protein. JImmunol 138:2547—2553, 1987

38. DULAWA J, RAMBAUSEK M, JANN K, NOTOHAMIPRODJO M, RITZE: Abnormal radiofurosemide binding by Tamm-Horsfall proteinof diabetic patients. Diabetologia 28:827—830, 1985

39. ALLEN F, TI5HER CC: Morphology of the ascending thick limb ofHenle. Kidney mt 9:8—22, 1976

40. KAISSLING B, PETERS 5, KRIZ W: The transition of the thickascending limb of Henle's loop into the distal convoluted tubule inthe nephron of the rat kidney. Cell Tissue Res 182:111—118, 1977

41. SIKRI KL, FOSTER CL, MACHUGH N, MARSHALL RD: Localisa-tion of Tamm-Horsfall glycoprotein in the human kidney usingimmunofluorescence and immunoelectron microscopical tech-niques.JAnat 132:597—605, 1981

42. KUMAR 5, JASANI B, HUNT JS, MOFFAT DB, ASSCHER AW: Asystem for accurate immunolocalization of Tamm-Horsfall proteinin renal biopsies. Histochem J 17:1251—1258, 1985

43. AVIS PJG: The development of a radioimmunoassay procedure forthe estimation of Tamm-Horsfall glycoprotein in human serum.C/in Sci Mo! Med 52:183—191, 1977

44. HUNT JS, PEACH RJ, BRUNISHOLZ MC, LYNN KL, MCGIVENAR: A sensitive and specific ELISA using a monoclonal captureantibody for detection of Tamm-Horsfall urinary glycoprotein inserum. J Immuno/ Meth 91:35—43, 1986

45. ZALC B, COLLET A, MONGE M, OLLIER-HARTMANN MP, JACQUEC, HARTMANN L, BAUMANN NA: Tamm-Horsfall protein, akidney marker is expressed on brain sulfogalactosylceramidepositive astroglial structures. Brain Res 291:182—187, 1984

46. CELlo MR: Renin containing cells in kidney transplants into theanterior eye chamber. Kidney mt 29:1234—1234, 1986

47. HOYER JR, RESNICK JS, MICHAEL AF, VERNIER RL: Ontogeny ofTamm-Horsfall urinary glycoprotein. Lab Invest 30:757—761, 1974

48. SIKRI KL, FOSTER CL, ALEXANDER DP, MARSHALL RD: Local-ization of Tamm-Horsfall glycoprotein in the fetal and neonatalhamster kidney as demonstrated by immunofluorescence andimmunoelectron microscopical techniques. Biol Neonate 39:305—312, 1981

49. JEFFCOATE TNA, SCOTT JS: Polyhydramnios and oligohydram-nios. Can Med Assoc J 3 3:77—86, 1959

50. Ross N, MAZZUCHI N, PERCAROVICH R, RODRIGUEZ I, SANGUI-NETTI CM: Identification of Tamm-Horsfall urinary glycoproteinin human amniotic fluid. Am J Obsiex Gynecol 122:790—791, 1975

51. MEBERG A, HAUGEN H, AKE550N I, SANDE H: Uromucoid(Tamm-Horsfall's mucoprotein) in amniotic fluid and in urine inchildren. Nephron 23:28—31, 1979

52. PHIMISTER GM, MARSHALL RD: Tamm-Horsfall glycoprotein inhuman amniotic fluid. C/in Chim Acta 128:261—269, 1983

53. GREVEN J, KOLLING B, BRONEWSKI-SCHWARZER B, JANKER M,NEFFGEN B, NILIUS R: Evidence for a role of the Tamm Horsfallprotein in the tubular action of furosemide-like loop diuretics, inDiuretics: Chemistry, Pharmacology and Clinical Applications,edited by PUSCHETT JP, GREENBERG A, New York, Elsevier,1984, pp. 203—214

54. BRUNISHOLZ MC, LYNN KL, HUNT JS: Loop-acting diuretics donot bind to Tamm-Horsfall urinary glycoprotein. Clin Sci 73:305—310, 1987

55. Buii-St.n-oso AW, SCOTT DM, KINNE R: Localization ofTamm-Horsfall protein in chloride transporting epithelia: Lack ofcorrelation with the Na, K, Cl cotransporter. Eur J Cel/ Biol43:104—109, 1987

56. ORSKOV I, FERENCZ A, ORSKOV F: Tamm Horsfall protein oruromucoid is the normal urinary slime that traps type I fimbriatedEscherichia co/i. Lancet 1:887, 1980

57. ORSKOV I, FERENCZ A, ORSKOV F: Comparison of Escherichiacoli fimbrial antigen F7 with Type I fimbriae. Infect Immun27:657—666, 1980

58. O'HANLEY P. LARK D, FALKOw S, SCHOOLNIK G: Molecularbasis of Escherichia coli colonization of the upper urinary tract inBalb/c mice. J C/in Invest 75:347—360, 1985

59. KURIYAMA SM, SILVERBLATT FJ: Effect of Tamm-Horsfall uri-nary glycoprotein on phagocytosis and killing of type I fimbriatedEscherichia co/i. Infect Immun 51:193—198, 1986

60. SCHACHNER MS, MINITER PM, MAYRER AR, ANDRIDE VT:Interaction of Tamm-Horsfall protein with bacterial extracts.Kidney mt 3 1:77—84, 1987

61. MUCHMORE AM, DECKER JM: Uromodulin: an immunosuppres-sive 85 kD glycoprotein isolated from human pregnancy urine is ahigh affinity ligand for recombinant IL-l a. J Bio/ Chem 261:13404—13407, 1986

62. SI-IERBLOM A, DECKER J, MUCHMORE A: The lectin-like interac-tion between recombinant tumor necrosis factor and uromodulin.J Biol Chem 263:5418—5424, 1988

63. MUCHMORE A, DECKER J: Evidence that recombinant IL-I aexhibits lectin-like specificity and binds to homogeneous uromod-ulin via N-linked oligosaccharides. J Immuno/ 138:2541—2546,1987

64. SHERBLOM A, SATHYAMOORTHY M, DECKER JM, MUCHMOREAV: IL-2, a lectin with specificity for high-mannose glycopep-tides. J Immunol 143:939—944, 1989

65. MOONEN P, GAFFNER R, WINGFIELD P: Native cytokines do notbind to uromodulin (Tamm-Horsfall glycoprotein). FEBS Lett226:314—318, 1988

66. BALDWIN RL, CHANG MP, BRAMHALL J, GRAVES 5, BONAVIDA

Kumar and Muchmore: Tamm-Horsfall protein—uromodulin 1401

B, WIsNIEsKI BJ: Capacity of tumor necrosis factor to bind andpenetrate membranes is pH-dependent. J Immunol 141:2352—2357,1988

67. DOWER SK, KRONHEIM SR, MARCH CJ, CONLON PJ, Hops' TP,GILus S, URDAL DL: Detection and characterization of highaffinity plasma membrane receptors for human interleukin-l. JExp Med 162:501—515, 1985

68. MACDONALD HR, WINGFIELD P. SCHMEISSNER U, SHAW A,CLORE GM, GRONENBORN AM: Point mutations of human inter-leukin-1 with decreased receptor binding affinity. FEBS Lett209:295—298, 1986

69. DONAHUE JH, ROSENBERG SA: The fate of interleukin-2 after invivo administration. J Immunol 130:2203—2208, 1983

70. LOTZE MT, MATORY YL, ETTINGHAUSEN SE, RAYNER AA,SHARROW SO, SEIPP CA, CUSTER MC, ROSENBERG SA: In vivoadministration of purified human interleukin 2. II. Half life,immunologic effects, and expansion of peripheral lymphoid cellsin vivo with recombinant IL 2. J Immunol 135:2865—2875, 1985

71. NEWTON RC, UHL J, C0vINT0N M, BACK 0: The distribution andclearance of radiolabeled human interleukin-1 beta in mice. Lym-phokine Res 7:207—216, 1988

72. PESSINA GP, PAcINI A, B0CCI V, MAIOLI E, NALDINI A: Studieson tumor necrosis factor (TNF): II. Metabolic fate and distributionof human recombinant TNF. Lymphokine Res 6:35—44, 1987

73. BEUTLER BA, MILSARK I, CERAMI A: Cachetin/TNF: Production,distribution, and metabolic fate in vivo. Jlmmunol 135:3972—3977,1985

74. HUNT JS, MCGIVEN AR: Stimulation of human peripheral bloodlymphocytes by Tamm-Horsfall urinary glycoprotein. Immunol-ogy 35:391—395, 1978

75. FRANCESHI C, LICASTRO F, CHIRICOLO M, SERAFINI F, TABAC-CHI P: Monosaccharides and Tamm Horsfall glycopeptides inhibitallogeneic antigen induced lymphocyte blastogenesis in one-waymixed lymphocyte reaction, in Lectins—Biology, Biochemistry,Clinical Biochemistry (vol II), edited by BOG-HANSEN TC, Berlin,Walter de Gruyter, 1982, pp. 153—166

76. SEItxFINI-CEssI F, FRANCESCHI C, SPERTI 5: Specific interactionof human Tamm-Horsfall glycoprotein with leucoagglutinin, alectin from phaseolus vulgaris (red kidney bean). Biochem J183:381—388, 1979

77. DZIUKAS U, STERZEL RB, HODSON CR, HOYER JR: Renallocalisation of Tamm-Horsfall protein in unilateral obstructiveuropathy in rats. Lab Invest 47:185—193, 1982

78. HOWIE AJ, BREWER DB: Extra-tubular deposits of Tamm-Hors-fall protein in renal allografts. J Pathol 139:193—206, 1983

79. COHEN AH, BORDER WA, RAJFER J, DUMKE A, GLASSOCK RJ:Interstitial Tamm-Horsfall protein in rejecting renal allografts:Identification and morphologic pattern of injury. Lab Invest 50:519—525, 1984

80. RESNICK JS, Sisso 5, VERNIER RL: Tamm-Horsfall protein:Abnormal localization in renal disease. Lab Invest 38:550—555,1978

81. ZAGER RA, COTRAN RS, HOYER JR: Pathologic localization ofTamm-Horsfall protein in interstitial deposits in renal disease. LabInvest 38:52—57, 1978

82. CHAMBERS R, GROUFSKY A, HUNT JS, LYNN KL, MCGIVEN AR:Relationship of abnormal Tamm-Horsfall glycoprotein localizationto renal morphology and function. Clin Nephrol 26:21—26, 1986

83. MACKENZIE R, ASSCHER AW: Progression of chronic pyelonephri-tis in the rat. Nephron 42:171—176, 1986

84. HOYER JR: Tubulointerstitial immune complex nephritis in ratsimmunized with Tamm-Horsfall protein. Kidney mt 17:284—292,1980

85. FRIEDMAN J, HOYER J, SElLER MW: Formation and clearance oftubulointerstitial immune complexes in kidneys of rats immunizedwith heterologous antisera to Tamm-Horsfall protein. Kidney Int21:575—582, 1982

86. MAYRER AR, KASHGARIAN M, RIJDDLE NH, MARIER R, HODSON

C, RICHARDS F, ANDRIOLE VT: Tubulointerstitial nephritis andimmunologic responses to Tamm-Horsfall protein in rabbits chal-lenged with homologous urine or Tamm-Horsfall protein. J Immu-nol 128:2634—2642, 1982

87. BERKE ES, MAYRER AR, MINITER P, ANDRIOLE VP: Tubuloin-terstitial nephritis in rabbits challenged with homologous Tamm-Horsfall protein: The role of endotoxin. Clin Exp Immunol 53:562—572, 1983

88. GRANT AMS, BAKER LRI, NEUBERGER A: Urinary Tamm-Hors-fall glycoprotein in certain kidney diseases and its content in renaland bladder calculi. Clin Sci 44:377—384, 1973

89. GJALDBAEK JC, ROBERTSON WG: Does urine from stone formerscontain macromolecules which promote the crystal growth rate ofcalcium oxalate crystals in vitro? Clin Chim Acta 108:75—80, 1980

90. KITAMURA T, PAK CYC: Tamm and Horsfall glycoprotein doesnot promote spontaneous precipitation and crystal growth ofcalcium oxalate in vitro. J Urol 127:1024—1026, 1982

91. SCURR DS, LATIF AB, SERGEANT V, ROBERTSON WG: Polyani-onic inhibitors of calcium oxalate crystal agglomeration in urine.Proc EDTA 20:440—444, 1983

92. GAMBARO G, BAGGIO B, FAVARO 5, CICERELLO E, MARCHINI F,BORSATTI A: role de Ia mucoprotein de Tamm-Horsfall dans Ialithogenese oxalique-calcique. Neprhologie 5:171—172, 1984

93. SCURR DS, ROBERTSON WG: Modifiers of calcium oxalate crys-tallization found in urine. III. Studies on the role of Tamm-Horsfall mucoprotein and of ionic strength. J Urol 136:505—507,1986

94. HALLSON PC, ROSE GA: Uromucoids and urinary stone forma-tion. Lancet 1:1000—1002, 1979

95. ROSE GA, SULAIMAN 5: Tamm-Horsfall mucoproteins promotecalcium oxalate crystal formation in urine: Quantitative studies. JUrol 127:177—179, 1982

96. SOPHASAN 5, CHATASINGH S, TI-IANAPHAICHITR P, DHANAMITTA5: Tamm Horsfall mucoprotein in urine of potential bladder stoneformers. J Urol 124:522—524, 1980

97. WORCESTER E, NAKAGAWA Y, WABNER CL, KUMAR 5, COE FL:Crystal adsorption and growth slowing by nephrocalcin, albumin,and Tamm-Horsfall protein. Am J Physiol 255:F1 197—F1205, 1988

98. KEUTEL HJ: Immunobiologische und biochemische Untersuchun-gen der Harnkolloide. Em Beitrag zur Harnsteingenese. Ann UnivSaraviensis, Med IX. Saarbrucken., 1961

99. VERMOOTEN V: The occurence of renal calculi and their possiblerelation to diet. JAm Med Assoc 109:857—859, 1937

100. HOSTETTER TH, BRENNER BM: Renal circulatory and nephronfunction in experimental acute renal failure, in Acute RenalFailure (2nd ed), edited by BRENNER BM, LAZARUS JM, NewYork, Churchill Livingston, 1988, pp. 76—77

101. BzIs M, ROSEN 5, EPSTEIN FH: Acute renal failure, in TheKidney, 3rd edition, edited by BRENNER BM, RECTOR FC, Phila-delphia, WB Saunders Co, 1986, p. 749

102. COHEN AH, BORDER WA: Myeloma kidney: An immunomor-phogenetic study of renal biopsies. Lab invest 42:248—256, 1980

103. BERDON WE, SCHWARTZ RH, BECKER J, BAKER DH: Tamm-Horsfall proteinuria: Its relationship to prolonged nephrogram ininfants and children and to renal failure following intravenousurography in adults with multiple myeloma. Radiology 92:714-.722, 1969

104. BYRD L, SHERMAN RL: Radiocontrast induced acute renal failure.Medicine 58:270-279, 1979

105. EISENBERG RL, BANK WO, HEDGOCK MW: Renal failure aftermajor angiography can be avoided with hydration. AJR 136:859—861, 1981

106. GARNICK MB, MAYER RJ, ABELSON HT: Acute renal failureassociated with cancer treatment, in Acute Renal Failure (2nd ed),edited by BRENNER BM, LAZARUS JM, New York, ChurchillLivingston, 1988, pp. 636—645

107. RoMoLo JL, GOLDBERG NH, HANDE KR, ROSENBERG SA: Effectof hydration on plasma methotrexate levels. Cancer Treat Rep61:1393—1396, 1977