Division of Clinical Nephrology and Rheumatology, Niigata University Graduate School of Medical and Dental Sciences, Niigata, Japan.
Correspondence to Dr. Minoru Sakatsume, Division of Clinical Nephrology and Rheumatology, Niigata University Graduate School of Medical and Dental Sciences, 1-757 Asahimachi-dori, Niigata 951-8510, Japan. Phone: 81-25-227-2200; Fax: 81-25-227-0775; E-mail: sakatsum{at}med.niigata-u.ac.jp
ABSTRACT. Leukocyturia is associated with postinfectious glomerulonephritis(GN), interstitial nephritis, and renal allograft rejection.In addition, prominent infiltration of T cells and macrophagesis commonly observed in the renal tissues of patients with GN,accompanied by cellular crescent formation and/or interstitialcell infiltration. Because these infiltrating T cells were thoughtto participate in the development of the diseases and to appearin the urinary space while functioning as effector cells inthe renal inflammatory lesion, the study focused on the characterizationof T cells that appeared in urine. Freshly voided urine cellswere analyzed by flow cytometry to determine their phenotypeand by reverse transcriptase–PCR to detect cytokine mRNA.In urine from patients with different forms of GN, includingIgA nephropathy, Henoch-Schönlein purpura nephritis, andanti-neutrophil cytoplasmic antibody-associated GN, T cellsappeared together with macrophages. The urine T cells were mainlyCD45RA-, CD45RO+, and CD62L (L-selectin)-, which are the phenotypicfeatures of effector T cells. In agreement with this finding,T cells infiltrating glomeruli, crescents, and tubulointerstitiallesions were also effector type. Moreover, these urine cellsexpressed mRNA of the T helper lymphocyte 1 cytokines, interleukin-2,and/or interferon-. These findings suggest that the appearanceof effector T cells in urine may reflect the cellular immunereaction that occurs in the kidneys of patients with GN accompaniedby active cell infiltration.
Bacterial infection of the urinary tract is the most commoncause of leukocyturia. However, with some inflammatory glomerulardiseases, such as postinfectious glomerulonephritis (GN) andlupus nephritis, leukocytes are also found in urine (1,2). Moreover,the appearance of eosinophils or lymphocytes in urine has diagnosticsignificance for drug-induced interstitial nephritis (1) orrenal transplant allograft rejection (3–5), respectively.Thus, a portion of infiltrating leukocytes in glomeruli or tubulointerstitialinflammatory lesions, which may participate in the developmentof diseases, appear in urine.
It is very common for mononuclear cells, including T cells andmonocytes/macrophages, to be found as infiltrates in biopsyspecimens of patients with GN accompanied by crescent formationand/or tubulointerstitial cell infiltration (6,7). Immune responsesmediated by T cells and macrophages can be divided into twotypes on the basis of the expression of groups of T cell–derivedcytokines. A cellular immune response dominated by a T helperlymphocyte 1 (Th1) infiltration, in which interferon- (IFN-)and interleukin-2 (IL-2) (Th1 cytokines) predominate, generallyevokes a delayed-type hypersensitivity reaction. Immune responsesregulated by T helper lymphocyte 2 (Th2) cells result in thesecretion of IL-4, IL-5, and IL-10 (Th2 cytokines), which areessential for humoral immunity (8,9). It has been appreciatedthat Th1 cytokines are predominantly involved in the developmentof murine models of crescentic anti-glomerular basement membraneGN (10,11). However, it remains unclear whether the Th1 immuneresponse dominates in human GN accompanied by active cell infiltrationsuch as cellular crescents.
Human naive (virgin) and memory/effector T cells can be identifiedby the reciprocal expression of the CD45RA or CD45RO isoforms(12). This CD45RA/RO conversion occurs in lymphoid tissues orinflammatory lesions when T cells are activated after recruitmentfrom the general circulation, which is mediated by sets of homing/recirculationreceptors such as CD62L (L-selectin) (13–15). Naive Tcells express both CD45RA and CD62L. After stimulation by specificantigens, they lose CD45RA and acquire CD45RO along with bimodalexpression of CD62L. This differential regulation of CD62L expressionis mediated by distinct sets of cytokines. When naive T cellsare stimulated, it has been shown in vitro that IL-2, a Th1cytokine, induces downregulation of CD62L, whereas IL-6 andtransforming growth factor–ß1 promote its up-regulation(13). Recently, CD62L expression has been delineating functionalsubsets within the memory/effector T cell pool (16,17); CD62Lmemory T cells function as effectors. In human T cells, it wasreported (17) that a subset of CD62L memory T cells that arealso negative for expression of the chemokine receptor CCR7(which is also recognized as a functional T cell marker [(18])display immediate effector function.
Here we show that T cells that appear in urine of patients withGN accompanied by active cellular infiltration are mainly ofthe effector-type (CD62L-CD45RO+), which is the same as thephenotype of infiltrating T cells in and around glomeruli. Wealso demonstrate that these urinary T cells express Th1 cytokines.In the absence of bacterial infection along the urinary tract,the appearance of the effector T cells in urine might reflectthe cellular immune responses in inflammatory renal lesions.The analysis of urinary T cells from patients with GN couldprovide important clues to understanding the mechanism of thedevelopment of human GN.
Patients and Histologic Evaluation
Seventy-six Japanese individuals, either patients with persistenthematuria and/or proteinuria or healthy volunteers (n = 15),were examined. Patients with GN, including IgA nephropathy (IgAN)(n = 29), Henoch-Schönlein purpura nephritis (n = 4), anti-myeloperoxidaseanti-neutrophil cytoplasmic antigen (ANCA)–associatedGN (n = 10), idiopathic crescentic GN (n = 1), non-IgA mesangialproliferative GN (n = 3), membranous nephropathy (n = 6), minimalchange nephrotic syndrome (n = 3), and minor glomerular abnormality(n = 5), were diagnosed by renal biopsy. All patients were hospitalizedat Niigata University Hospital or affiliated hospitals afterbeing referred by physicians from outside our institution. Patientsgave informed consent for all studies. The patients with GNare the first 61 consecutive patients who consented to the studyand satisfied the conditions for the analyses described below.Urine and blood samples for flow cytometry and/or cytokine mRNAanalysis were collected during the week preceding the renalbiopsy. Renal biopsies were performed by the ultrasound-guidedpercutaneous needle biopsy method before immunosuppressive therapybegan. With the use of the biopsy specimens stained by periodicacid–Schiff stain, the grade of distribution of cellularcrescents or interstitial cell infiltration in the biopsy specimenswas designed as follows: 0, absent (or 4% of glomeruli or theinterstitial space evaluated); 1, scattered (5% to 24%); 2,focal (25% to 49%); and 3, extensive (50% to 74%) or diffuse(75%). The grade of severity of diffuse endocapillary cell infiltrationwas evaluated as follows: 0, absent; 1, mild; 2, moderate; and3, severe. In this study, there was no biopsy specimen thatwas graded as 3 for diffuse endocapillary cell infiltration.Two pathologists independently scored the microscopy samplesin a blinded manner.
Flow Cytometric Analysis of Urinary Cells and Peripheral Blood Mononuclear Cells
Fresh urine samples (50 ml) were centrifuged (1500 x g) at 4°C.The pellets were washed once with phosphate-buffered saline(PBS), resuspended with 100 µl of staining buffer (3%fetal bovine serum and 0.05% sodium azide in PBS) and dividedinto five tubes (Falcon 2052, Becton Dickinson, Lincoln Park,NJ) with 20 µl of the cell suspension in each. Stainingof cells was carried out as described elsewhere (19). Briefly,cell suspensions were incubated with 10 µl of FITC- and/orphycoerythrin–conjugated monoclonal antibodies (anti-CD3[UCHT1, mouse IgG1,], anti-CD62L [Dreg56, mouse IgG1,], anti-CD45RO[UCHL1, mouse IgG2a,], anti-CD45RA [HI100, mouse IgG2b,], anti-CD14[M5E2, mouse IgG2a,], anti-CD4 [RPA-T4, mouse IgG1,], and anti-CD8[RPA-T8, mouse IgG1,]) (PharMingen, San Diego, CA) for 30 minat 4°C. Isotype matched FITC- or phycoerythrin-conjugatednonspecific antibodies (mouse IgG1,; IgG2a,; or IgG2b,; PharMingen)were also used for negative controls. After incubation, 1 mlof hemolysis buffer (0.83% NH4Cl in 20 mM Tris-HCl buffer) wasadded, and the cells were incubated at 37°C for 3 min andwashed twice with 3 ml of staining buffer. After centrifugation,the pellets were suspended in 0.5 ml staining buffer. The cellsuspensions were stored on ice and shaded from light until fluorescence-activatedcell sorter (FACS) analysis. Two-color flow cytometric analysiswas carried out on a FACScan (Becton Dickinson, Franklin Lakes,NJ) with Cellysis software (Becton Dickinson). Dead cells wereexcluded by forward scatter, side scatter, and propidium iodidegating. The gates for mononuclear cells (G1) and lymphocyticcells (G2) were set according to methods published elsewhere(20,21). These gates were first set for analyzing peripheralblood cells, then urine cells were examined under the same setting.Counts of T lymphocytes or monocytes/macrophages were determinedby multiplying the number of viable cells in the gated mononuclearcell-region in each sample by the percentage of CD3+ or CD14+cells in the population. Because the number of cells in urinemay be affected by the state of urinary dilution or concentration,the counts of cells were standardized by the osmotic pressureof the urine; isotonic pressure, 280 mOsm/kgL, was designatedas 1.
Peripheral blood mononuclear cells were separated from heparinizedblood samples with a ficoll-gradient method by use of Lymphoprep(Nycomed Pharma AS, Oslo, Norway) and then washed twice withPBS. A total of 106 cells were suspended in 20 µl of stainingbuffer and incubated with antibodies as above. The subsequenttreatment and analysis were performed in the same manner asfor the urine samples. The cases who showed more neutrophilsthan mononuclear cells in urine by FACS analysis or a significantnumber of bacteria (>1 x 104 colonies/ml urine) in culturesof urine were excluded because of the possibility that bacterialinfection in the urinary system might be superimposed on glomerularinflammation.
Immunohistochemistry of Renal Biopsy Specimens
Sequential frozen sections (2 µm) of biopsy specimensfrom patients whose urine and blood had been sampled for FACSanalysis were treated with anti-CD3 (UCHT1), antiCD45RO (UCHL1),anti-CD14 (M5E2), or anti-CD62L (Dreg56) (Pharmingen) as primaryantibodies. Subsequently, sections were treated with goat anti-mouseIg-conjugated peroxidase-labeled polymer (Dako EnVision, Dako,Carpinteria, CA), and the colorimetric reaction was developedby use of diaminobenzidine (Dako). Counterstaining was performedwith hematoxylin.
Detection of Specific mRNA
Peripheral blood mononuclear cells from a healthy volunteer,separated from heparinized blood samples as mentioned above,were incubated on dishes precoated with human IgG and rabbitanti-human IgM to remove monocytes and B cells, as describedelsewhere (22). The nonadherent cells contained 91% CD3+ T cellsas verified by FACS analysis. These T cell–enriched mononuclearcells were suspended in RPMI 1640 medium supplemented with 10%normal human serum and L-glutamine. Then 107 cells were incubatedin 2 ml of medium per well in 6-well plates with phorbol 12-myristate13-acetate (5 ng/ml) and ionomycin (1 µM) (Sigma, St.Louis, MO) for 8 h. These samples were used as positive controlsfor cytokine expression. The stimulated T cells (positive control)and urinary cells from the 50 to 200 ml of freshly voided urinewere washed twice with ice-cold PBS and pelleted by centrifugation.Total RNA was prepared from cells by use of Isogen (Nippon Gene,Tokyo, Japan), following the manufacturer’s protocol.Single-stranded cDNA was synthesized from 1 µg of totalRNA in a 20-µl reaction mixture that contained 5 µMrandom hexamer and 1 U reverse transcriptase (Superscript II,Life Technology, Rockville, MD). The indicated amount of reactionmixture was then subjected to PCR for 30 cycles as describedelsewhere (23): 1 min at denaturation at 94°C, 1 min ofannealing at 60°C, and 1 min of extension at 72°C. Specificprimers used for PCR were human IL-2 (sense, 5'-ATGTACAGGATGCAACTCCTGTCTT-3';anti-sense, 5'-GTCAGTGTTGAGATGATGCTTTGAC-3'), human IL-4 (sense,5'-ATGGGTCTCACCTCCCAACTGCT-3'; anti-sense, 5'-CGAACACTTTGAATATTTCTCTCTCAT-3'),human IFN- (sense, 5'-ATGAAATATACAAGTTATATCTTGGCTTT-3'; anti-sense,5'-GATGCTCTTCGACCTCGAAACAGCAT-3') (24), and human T cell receptorß-chain common region (TCR-Cß) (sense, 5'-CCCACACCCAAAAGGCCA-3';anti-sense, 5'-CATAGAGGATGGTGGCAG-3'). For detection of IL-4mRNA, a subsequent nested PCR (25 cycles) was carried out with1/100 of the PCR product with the use of the following primers(sense, 5'-CAACTGCTTCCCCCTCTGTTCT-3'; anti-sense, 5'-CTCTCTCATGATCGTCTTTAG-3')(24) with the same regimen of amplification as the first round.To visualize specific PCR products, Southern blot analysis wasperformed after running 10 µl of the samples on 2% agarosegels as described elsewhere (22). Oligonucleotide probes usedwere 5'-TACATTTAGTAATCTAGCTGGA-3' for IL-2, 5'-ACAAGTTATATCTTGGCTTTTCAGCTCT-3'for IFN-, 5'-CTGCTAGCATGTGCCGGCAACT-3' for IL-4, and 5'-CAATGACTCCAGATACTGCCT-3'for TCR-Cß. Semiquantitation of TCR-Cß mRNAwas performed by following a method published elsewhere (25,26).
Statistical Analyses
An independent nonparametric test (Mann-Whitney U test) wasused to determine the significance of differences between groups.The data are presented as mean ± SEM. The correlationbetween the grade of cell infiltration in renal tissues andthe number of T cells and macrophages in urine was assessedby Spearman rank correlation test. The rs represents the Spearmancorrelation coefficient.
Phenotype of Urine T Cells
As shown in Table 1, patients with minor glomerular abnormality,minimal change nephrotic syndrome, membranous nephropathy, ornon-IgA mesangial proliferative GN, whose renal biopsy specimenswere not accompanied by cell infiltration, such as cellularcrescent formation, interstitial cell infiltration, or endocapillarycell infiltration, showed either no or a small number (<47/ml)of T cells (CD3+ cells) and macrophages (CD14+ cells) in urine,which was similar to the data from normal control subjects.In contrast, relatively high numbers of T cells and macrophageswere detected in the urine of patients with glomerular diseasesaccompanied by active cellular infiltrates, such as Henoch-Schönleinpurpura nephritis, anti-myeloperoxidase ANCA-associated GN,idiopathic crescentic GN, and a portion of IgAN. When the relationshipbetween the number of T cells and macrophages in urine and thegrade of cell infiltration in renal tissues was assessed (Figure 1),strong correlations existed between that number and thegrade of cellular crescent formation (rs = 0.746, P < 0.0001)or the grade of interstitial cell infiltration (rs = 0.752,P < 0.0001), regardless of histologic diagnosis. A moderatecorrelation also existed between that number and the grade ofendocapillary cell infiltration (rs = 0.432, P < 0.001).Some cases with IgAN were exceptional in that a number of Tcells and macrophages appeared in the urine (100/ml, for example,for patient 3 in Table 2), even though the inflammatory cellswere not observed in renal tissues. However, some form of cellinfiltration was observed in renal tissues in all patients whoshowed >120 T cells and macrophages/ml of urine.
Figure 1. The relationship between the number of T cells and macrophages in urine and the grade of forms of cell infiltration in renal tissues. The grading of distribution of cellular crescents or interstitial cell infiltration and the grading of severity of diffuse endocapillary cell infiltration in the biopsy specimens were performed as described in Materials and Methods. The horizontal lines in each column represent the 50th percentile, and the upper and lower dotted horizontal lines represent the 75th and 25th percentiles, respectively. rs represents Spearman correlation coefficient.
To determine whether these T cells in urine were derived frominflamed renal lesions or were merely released from rupturedglomerular capillaries or other damaged renal vessels, we comparedthe phenotype of urine T cells with that of peripheral bloodT cells, focusing on naive/memory markers such as CD45RA (anaive T cell marker), CD45RO (a memory T cell marker), and CD62L(L-selectin). Patients who showed a sufficient number of T cellsin urine for detailed analysis (>30 T cells/ml) (n = 22)were examined further and listed in Table 2. Urine T cells hadlow CD62L expression (urine T cells versus peripheral bloodT cells: 18.4 ± 2.0% versus 65.2 ± 2.0%, P <0.0001), high CD45RO expression (80.8 ± 3.0% versus 50.0± 4.2%, P < 0.0001), and low CD45RA expression (15.1± 1.5% versus 52.0 ± 4.0%, P < 0.0001) as determinedby the frequency of positive cells when compared with the peripheralcirculating T cells of these patients (Table 2). There was alsoa significant difference between the CD4/CD8 ratio of urineT cells and peripheral blood T cells (0.9 ± 0.1 versus2.1 ± 0.6, P < 0.01) (Table 2). Flow cytometric datafrom a representative patient (12 [with Henoch-Schönleinpurpura nephritis] in Table 2) are shown in Figure 2. Theseresults indicate that the appearance of urine T cells may beassociated with GN accompanied by active cell infiltration andthat the majority of the urine T cells are CD62L-CD45RO+ effectorT cells.
Figure 2. Two-color flow cytometric analysis of urine T cells. Fresh urine cells (A) and peripheral blood mononuclear cells (B) were stained with FITC- or phycoerythrin (PE)-conjugated monoclonal antibodies (anti-CD3, anti-CD62L, anti-CD45RO, anti-CD45RA, or anti-CD14). Cells in the gated area G2 (lymphocyte region) of forward/side scatter profiles (a, g) were analyzed, and data are shown in b, c, d, e and h, i, j, k. Analyses of cells in the gate G1 (mononuclear cell region) are shown in f and i. A negative control of the gate G2 by use of FITC- or PE-conjugated nonspecific antibodies (mouse IgG1,) is also shown (b, h).
Infiltration of Effector T Cells in Renal Tissues
Because the majority of urine T cells of patients with GN accompaniedby active cellular infiltrates were CD62L-CD45RO+, we examinedwhether T cells that infiltrated into the renal tissues of thesepatients also had the same phenotype. As shown in Figure 3,which was derived from the biopsy specimen of patient 18 (Table 2;anti-myeloperoxidase ANCA–associated GN), T cells (CD3+)were predominant as infiltrating mononuclear cells in glomeruliand periglomerular lesions, and most T cells were positive forCD45RO. CD14+ cells were also present around the lesions, whereasCD62L+ cells were rarely observed. In some instances, CD62Lwas expressed on infiltrating T cells in lymph follicle lesions,which were observed in the corticomedullary junction (Figure 3E).These findings suggest that the T cells in urine may bepredominantly derived from the effector T cells in the inflamedrenal tissues and that these T cells may provide informationabout inflammatory events that are occurring in the kidneys.
Figure 3. Immunohistochemistry of renal biopsy specimens. Sequential sections of biopsy specimens from patient 18 (Table 2; antineutrophil cytoplasmic antigen (antimyeloperoxidase)–associated glomerulonephritis) who showed diffuse crescent formation and interstitial cell infiltration around glomeruli, were stained with anti-CD3 (A), anti-CD45RO (B), anti-CD14 (C), and anti-CD62L (D). T cells (CD3+) are predominant as infiltrating mononuclear cells in glomeruli, crescents, and the periglomerular lesions (A), and most T cells are positive for CD45RO (B). CD14+ cells are also present around the lesion (C). The infiltrating mononuclear cells rarely express CD62L (D, left), whereas, in lymph follicle lesions of the same biopsy specimen, T cells are positive for CD62L (D, right). IL, interleukin; IFN, interferon. Magnification, x200.
Cytokine mRNA Expression in Urine Cells
To determine the functional role of the urine T cells, totalRNA was prepared from urine cells collected from patients whoshowed many urine T cells by flow cytometry and subjected toreverse transcriptase–PCR analysis to detect mRNA of Th1(IL-2 and IFN-) and Th2 (IL-4) cytokines. As a positive controlfor cytokine expression, peripheral blood lymphocytes from ahealthy volunteer were stimulated with phorbol 12-myristate13-acetate (5 ng/ml) and ionomycin (1 µM) for 8 h, andtotal RNA was prepared. In the urinary cells, the populationof T cells bearing ß-TCR was the major subpopulationin the lymphocyte gate (>80%). Because T cells are thoughtto be a major source for expression of the cytokines IL-2, IFN-,and IL-4 (27–29), TCR-Cß was amplified in cDNAsamples to determine the frequency of ß-T cells. ThePCR products were hybridized with 32P-labeled internal Cßprobe, and the radiointensity was measured by a radio-imager(Bas-2000, Fuji Film, Tokyo, Japan). The values for the intensityof the bands for the PCR products of TCR-Cß from thepositive control were plotted in a log-log graph against theamount of RNA to obtain a standardization curve, where a linearrelationship was obtained. The amount of RNA from T cells ineach sample was then estimated by use of this curve (Figure 4A).Equivalent amounts of ß-T cell–derivedRNA (103 pg) from each sample were subjected to PCR and subsequentSouthern blot analysis to detect mRNA expression of the cytokines,IL-2, IFN-, or IL-4. Samples that showed either low TCR-CßmRNA expression or an insufficient amount of cDNA for furtherPCR analysis were omitted. All samples of urine cells expressedmRNA of either IL-2 or IFN-, and most of these samples expressedmRNA of both cytokines, whereas expression of IL-4 mRNA wasnot detected (Figure 4B). Thus, urine T cells, which were thoughtto be responsible for the expression of these cytokines, expressTh1 cytokines at the mRNA level, although the number of caseswe could examine was limited.
Figure 4. Semiquantitation of human T cell receptor ß-chain common region (TCR-Cß) mRNA and detection of cytokine mRNA from urine T cells by reverse transcriptase (RT)–PCR. (A) TCR-Cß mRNA was amplified by RT-PCR by use of total RNA from control peripheral blood lymphocytes (PBL) stimulated by phorbol 12-myristate 13-acetate and Ca-ionophore and from urine cells as described in Materials and Methods. In the upper panel, 5 x 102 (a), 1.5 x 103 (b), and 4.5 x 103 (c) pg RNA of PBL from a healthy volunteer were subjected to RT-PCR with TCR-Cß primers. The same PCR analysis was also simultaneously performed with RNA samples (0.1 µg) from urine cells of these patients. The numbers 2, 5, 6, 8, 10, 11, 16, and 17 correspond to patient numbers listed in Table 2. The PCR products were electrophoresed on 2% agarose gel and hybridized with a specific 32P-labeled internal probe for TCR-Cß after transferring to a nylon membrane. The radio-intensities were quantitated by Bas-2000 (Fuji film), and data were plotted on a log-log graph, as shown in the lower panel, where the specific PCR products were shown to be linearly amplified under these conditions. (B) Equivalent amounts of ß-T cell–derived RNA (103 pg) from patients, which had been estimated in panel A, were subjected to RT-PCR analysis to detect IL-2, IFN-, and IL-4 mRNA, as well as TCR-Cß mRNA. The same sets of RT-PCR were also performed with samples of activated PBL from a healthy volunteer (0.5 x 103 [a], 1.5 x 103 [b], and 4.5 x 103 [c] pg RNA).
This study demonstrated that effectors of cell-mediated immunity,T cells and macrophages, appear in the urine of patients withGN accompanied by prominent cellular infiltrates and that theT cells express the immediate effector-phenotype and Th1 cytokines.The urinary T cells were found to be mainly effector T cells(CD45RO+CD62L-). The majority of infiltrating T cells in renalbiopsy specimens of these patients, which were found in glomeruliand periglomerular interstitial lesions, were also of the effectortype (CD45RO+CD62L-). On the other hand, non–effectorT cells (CD62L+ cells) were accumulated in lymph follicle lesions,where T cell conversion from a naive cell to a memory/effector-typemight be occurring because of stimulation with kidney-specificantigens. The T cells may have lost CD62L during activation,presumably under the predominant influence of Th1 cytokinessuch as IL-2. Thereafter, those T cells may have functionedas effectors, participating in the development of kidney diseasesbefore appearing in the urine. It has recently been shown byCunningham et al. (30) that T cells and macrophages are prominentin glomeruli of ANCA-associated GN, which is driven by cell-mediatedimmunity, and that many effector (CD45RO+) T cells (73.0% ofthe T cells) are observed in those glomeruli. This finding correspondswell with our results (Table 2). Thus, the appearance of T cellsand macrophages in urine seems to reflect the cell-mediatedinflammatory events that are occurring in the kidneys.
The reason why a number of T cells and macrophages (about 100/ml)appeared in some patients with IgAN in the absence of cellularinfiltrates in the tissues might be related to the mechanismof development of IgAN itself or to the limitations of smallbiopsy specimens, which might miss localized lesions withinthe kidney. However, the appearance of >120 T cells and macrophages/mlin urine was indicative of the presence of cell infiltrationin renal tissues, even in IgAN. Regardless of histologic diagnosis,the number of T cells and macrophages in urine correlated withthe grade of cell infiltration in renal tissues, especiallycellular crescent formation and interstitial cell infiltration.Therefore, it is possible, to some extent, to predict how severelythe renal tissues are damaged by cellular immunity-mediatedinflammation by measuring the number of the T cells and macrophagesin urine.
The regulatory interactions of functionally distinct helper(or cytotoxic) T cell subsets, Th1 (or Tc1) and Th2 (or Tc2),are mediated by the cytokines they produce (8,9), resultingin Th1/Th2 polarization of cellular immune responses in immunopathologicdisorders (10,11,31–34). In this study, it has been suggestedthat urine T cells associated with cellular immunity-mediatedGN might be Th1-type or Tc1-type by cytokine mRNA analysis,although we could not exclude the possibility that not onlyT cells, but also natural killer cells or natural killer T cells,might participate in the expression of IL-2 and IFN- mRNA. Itwas somewhat unexpected that messages of only Th1 cytokineswere detected even in IgAN, an immune complex-mediated GN. Interestingly,it has been reported that, in active Heymann GN, although aTh2 immune response is thought to be crucial at the stage ofanti-Fx1A antibody production, actual glomerular injury is mediatedby Th1 immune cells (33–35). Th1 cellular immune responsesmay be responsible, in all forms of GN, for the tissue injurythat is mediated by cellular immunity. It is possible, however,that in allergic interstitial nephritis, which is known to beaccompanied by eosinophil infiltration (1) and was not includedin this study, a Th2 immune response could be dominant, andmRNA of Th2 cytokines could be detected in urine cells. Kaneganeet al. (36) reported that CD4+CD45RO+CD62L- human T cells preferentiallyproduce a Th1 cytokine, IFN-, but do not produce IL-4 or IL-5.This is consistent with our data on urine T cells, althoughit is not clear whether CD8+CD45RO+CD62L- T cells are also Tc1-type.
The urine T cells of GN accompanied by active cell infiltrationare presumed to have organized Th1 immune reactions as effectorsin the inflammatory lesions of kidneys by producing IL-2 andIFN-. The CD14+ macrophages, which appear in urine togetherwith T cells, may be activated by these T cells, especiallythrough the IFN- they produce, and also serve as effector cellsthat mediate tissue injury. Hotta et al. (37,38) showed thatCD16+CD14+ effector type-macrophages appear in the renal tissueand urine of patients with fresh crescentic GN and active IgAN.Thus, T cells and macrophages in urine may reflect a Th1-typeimmune reaction that is occurring in kidneys of GN with prominentcellular infiltrates and may have participated in the developmentof inflammation of kidneys before appearing in urine. Therefore,the detection of such T cells in urine may be highly suggestiveof the presence of active cellular infiltrates in the kidneys.On the basis of this study and our unpublished observations,three possible uses of this method can be considered: (1) whenpatients present rapidly progressive renal dysfunction, nephropathythat shows massive proteinuria, but no active cell infiltrationin renal tissues, such as minimal change disease, could be distinguishedfrom crescentic GN before obtaining histologic diagnosis; (2)the efficacy of therapy against GN accompanied by active cellinfiltrates could be monitored or exacerbations of GN couldbe detected without rebiopsy; (3) when hematuria is overt, itcould be determined whether it is due to simple bleeding fromkidney or urinary tract or whether it is associated with activelyflaring kidney lesions.
Acknowledgments
This work was supported by the Ministry of Education, Science,Sports, and Culture (grant-in-aid for Scientific Research [C]12670420 and [C] 13671104). The authors thank Satomi Takeuchi,Keiko Yamagiwa, and Naofumi Imai for their skillful technicalassistance. The authors also appreciate Dr. Hiroshi Kagamu andProfessor Kouhei Akazawa for helpful discussions.
Kasiske BL, Keane WF: Laboratory assessment of renal disease.In: The Kidney, 4th ed., edited by Brenner BM, Philadelphia, WB Saunders, 1996,p 1159
Kalia A, Travis LB: Manifestations of renal disease and renal insufficiency.In: Pediatric Kidney Disease, 2nd ed., edited by Edelmann CMJr, Boston, Little, Brown and Company, 1992,p 560
Toma H, Ishida Y, Takahashi K, Ota K: The clinical value of active T-rosette-forming cells in urine after renal transplantation. J Urol 1985; 133: 575–578,[Medline]
Lao M, Wyzgal J, Nowaczyk M, Gorski A: Value of urine sediment phenotyping in renal allograft recipients. Arch Immunol Ther Exp 41: 373–375, 1993
Roberti I, Reisman L, Burrows L, Lieberman KV: Urine cytology and urine flow cytometry in renal transplantation-a prospective double blind study. Transplant 59: 495–500, 1995[Medline]
Hooke DH, Gee DC, Atkins RC: Leukocyte analysis using monoclonal antibodies in human glomerulonephritis. Kidney Int 31: 964–972, 1987[Medline]
Nolasco FE, Cameron JS, Hartley B, Coelho A, Hildreth G, Reuben R: Intraglomerular T cells and monocytes in nephritis: Study with monoclonal antibodies. Kidney Int 31: 1160–1166, 1987[Medline]
O’Garra A, Murphy K: Role of cytokines in determining T-lymphocyte function. Curr Opin Immunol 6: 458–466, 1994[Medline]
Sad S, Marcotte RG, Mosmann TR: Cytokine-induced differentiation of precursor mouse CD8+ T cells into cytotoxic CD8+ T cells secreting Th1 or Th2 cytokines. Immunity 2: 271–279, 1995[Medline]
Kitching AR, Tipping PG, Huang XR, Mutch DA, Holdsworth SR: Interleukin-4 and interleukin-10 attenuate established crescentic glomerulonephritis in mice. Kidney Int 52: 52–59, 1997[Medline]
Kitching AR, Tipping PG, Mutch DA, Huang XR, Holdsworth SR: Interleukin-4 deficiency enhances Th1 responses and crescentic glomerulonephritis in mice. Kidney Int 53: 112–118, 1998[Medline]
Michie CA, McLean A, Alcock C, Beverley PC: Lifespan of human lymphocyte subsets defined by CD45 isoforms. Nature 360: 264–265, 1992[Medline]
Picker LJ, Treer JR, Ferguson DB, Collins PA, Buck D, Terstappen LW: Control of lymphocyte recirculation in man. I. Differential regulation of the peripheral lymph node homing receptor L-selection on T cells during the virgin to memory cell transition. J Immunol 150: 1105–1121, 1993[Abstract]
Jung U, Ramos CL, Bullard DC, Ley K: Gene-targeted mice reveal importance of L-selectin-dependent rolling for neutrophil adhesion. Am J Physiol 274: H1785–H1789, 1998[Abstract/Free Full Text]
Brady HR, Spertini O, Jimenez W, Brenner BM, Marsden PA, Tedder TF: Neutrophils, monocytes, and lymphocytes bind to cytokine-activated kidney glomerular endothelial cells through L-selectin (LAM-1) in vitro. J Immunol 149: 2437–2444, 1992[Abstract]
Ahmadzadeh M, Hussain SF, Farber DL: Heterogeneity of the memory CD4 T cell response: Persisting effectors and resting memory T cells. J Immunol 166: 926–935, 2001[Abstract/Free Full Text]
Sallusto F, Lenig D, Forster R, Lipp M, Lanzavecchia A: Two subsets of memory T lymphocytes with distinct homing potentials and effector functions. Nature 401: 708–712, 1999[Medline]
Sallusto F, Lanzavecchia A, Mackay CR: Chemokines and chemokine receptors in T-cell priming and Th1/Th2-mediated responses. Immunol Today 19: 568–574, 1998[Medline]
Sakatsume M, Harada Y, Ling X, Yasumoto A, Kurosu A, Koseki H, Saito T, Kantake M, Taniguchi M: Autoreactive T-cell clones which suppress cytotoxic T cell responses. Int Immunol 3: 377–384, 1991[Abstract/Free Full Text]
Thompson JM, Gralow JR, Levy R, Miller RA: The optimal application of forward and ninety-degree light scatter in flow cytometry for the gating of mononuclear cells. Cytometry 6: 401–406, 1985[Medline]
Salzman GC, Crowell JM, Martin JC, Trujillo TT, Romero A, Mullaney PF, LaBauve PM: Cell classification by laser light scattering: Identification and separation of unstained leukocytes. Acta Cytol 374–347, 1975
Sakatsume M, Finbloom DS: Modulation of the expression of the interferon- receptor beta-chain controls responsiveness to interferon- in human peripheral blood T cells. J Immunol 156: 4160–4166, 1996[Abstract]
Sakatsume M, Narita I, Yamazaki H, Saito A, Nakagawa Y, Kuriyama H, Kuwano R, Gejyo F, Arakawa M: Down-regulation of interferon- signaling by gene transfer of Stat1 mutant in mesangial cells. Kidney Int 57: 455–463, 2000[Medline]
Zaitseva MB, Golding H, Betts M, Yamauchi A, Bloom ET, Butler LE, Stevan L, Golding B: Human peripheral blood CD4+ and CD8+ T cells express Th1-like cytokine mRNA and proteins following in vitro stimulation with heat-inactivated Brucella abortus.Infect Immun 63: 2720–2728, 1995[Abstract]
Sumida T, Yonaha F, Maeda T, Tanabe E, Koike T, Tomioka H, Yoshida S: T cell receptor repertoire of infiltrating T cells in lips of Sjogren’s syndrome patients. J Clin Invest 89: 681–685, 1992
Choi YW, Kotzin B, Herron L, Callahan J, Marrack P, Kappler J: Interaction of Staphylococcus aureus toxin "superantigens" with human T cells. Proc Natl Acad Sci USA 86: 8941–8945, 1989[Abstract/Free Full Text]
Smith KA: Interleukin-2: Inception, impact, and implications. Science 240: 1169–1176, 1988[Abstract/Free Full Text]
Farrar MA, Schreiber RD: The molecular cell biology of interferon- and its receptor. Annu Rev Immunol 11: 571–611, 1993[Medline]
Paul WE, Ohara J: B cell stimulatory factor-1/interleukin 4. Annu Rev Immunol 5: 429–459, 1987[Medline]
Cunningham MA, Huang XR, Dowling JP, Tipping PG, Holdsworth SR: Prominence of cell-mediated immunity effectors in "pauci-immune" glomerulonephritis. J Am Soc Nephrol 10: 499–506, 1999[Abstract/Free Full Text]
Ito S, Ansari P, Sakatsume M, Dickensheets H, Vazquez N, Donnelly RP, Larner AC, Finbloom DS: Interleukin-10 inhibits expression of both interferon alpha– and interferon gamma–induced genes by suppressing tyrosine phosphorylation of STAT1. Blood 93: 1456–1463, 1999[Abstract/Free Full Text]
Sakatsume M, Igarashi K, Winestock KD, Garotta G, Larner AC, Finbloom DS: The Jak kinases differentially associate with the alpha and beta (accessory factor) chains of the interferon- receptor to form a functional receptor unit capable of activating STAT transcription factors. J Biol Chem 270: 17528–17534, 1995[Abstract/Free Full Text]
Penny MJ, Boyd RA, Hall BM: Role of T cells in the mediation of Heymann nephritis. II. Identification of Th1 and cytotoxic cells in glomeruli. Kidney Int 951: 1059–1068, 1997
Penny MJ, Boyd RA, Hall BM: Permanent CD8(+) T cell depletion prevents proteinuria in active Heymann nephritis. J Exp Med 188: 1775–1784, 1998[Abstract/Free Full Text]
Penny MJ, Boyd RA, Hall BM: Mycophenolate mofetil prevents the induction of active Heymann nephritis: Association with Th2 cytokine inhibition. J Am Soc Nephrol 9: 2272–2282, 1998[Abstract]
Kanegane H, Kasahara Y, Niida , Yachie A, Sughii S, Takatsu K, Taniguchi N, Miyawaki T: Expression of L-selectin (CD62L) discriminates Th1- and Th2-like cytokine-producing memory CD4+ T cells. Immunology 87: 186–190, 1996[Medline]
Hotta O, Taguma Y, Ooyama M, Yusa N, Nagura H: Analysis of CD14+ cells and CD56+ cells in urine using flow cytometry: A useful tool for monitoring disease activity of IgA nephropathy. Clin Nephrol 39: 289–294, 1993[Medline]
Hotta O, Yusa N, Ooyama M, Unno K, Furuta T, Taguma Y: Detection of urinary macrophages expressing the CD16 (Fc gamma RIII) molecule: A novel marker of acute inflammatory glomerular injury. Kidney Int 55: 1927–1934, 1999[Medline]
Received for publication October 30, 2000.
Accepted for publication June 5, 2001.
This article has been cited by other articles:
M. Sakatsume and F. Gejyo Effector T cells and macrophages in urine as a hallmark of systemic vasculitis accompanied by crescentic glomerulonephritis
Nephrol. Dial. Transplant.,
March 1, 2003;
18(3):
607 - 609.
[Full Text][PDF]
Top
Abstract
Introduction
. These findings suggest that the appearance of effector T cells in urine may reflect the cellular immune reaction that occurs in the kidneys of patients with GN accompanied by active cell infiltration. 
4% of glomeruli or the interstitial space evaluated); 1, scattered (5% to 24%); 2, focal (25% to 49%); and 3, extensive (50% to 74%) or diffuse (
75%). The grade of severity of diffuse endocapillary cell infiltration was evaluated as follows: 0, absent; 1, mild; 2, moderate; and 3, severe. In this study, there was no biopsy specimen that was graded as 3 for diffuse endocapillary cell infiltration. Two pathologists independently scored the microscopy samples in a blinded manner. 
], anti-CD62L [Dreg56, mouse IgG1,
100/ml, for example, for patient 3 in Table 2), even though the inflammatory cells were not observed in renal tissues. However, some form of cell infiltration was observed in renal tissues in all patients who showed >120 T cells and macrophages/ml of urine. 
ß-TCR was the major subpopulation in the lymphocyte gate (>80%). Because T cells are thought to be a major source for expression of the cytokines IL-2, IFN-
