Expression of the mucosal γδ T cell receptor V region repertoire in patients with IgA nephropathy

Kidney International, J/ol. 52 (1997), pp. 1047—1053

CLINICAL NEPHROLOGY - EPIDEMIOLOGY - CLINICAL TRIALS

Expression of the mucosal y T cell receptor V region repertoire in patients with IgA nephropathy COLLEEN OLIVE, ALICE C. ALLEN, STEVEN J. HARPER, ANTHONY C. B. WIcKs, JOHN FEEHALLY,

and MICHAEL C. FALK Renal Research Laborato,y, Department of Renal Medicine, Princess Alexandra Hospital Ipswich Road, Woolloongabba, Brisbane, Queensland, Australia, and Department of Nephrology, Leicester General Hospital Leicester, and Richard Bright Renal Unit, Southmead Hospital Bristol United Kingdom

Expression of the mucosal yl T cell receptor V region repertoire in patients with IgA nephropathy. IgA nephropathy (IgAN) is characterized by the deposition of IgA in the glomerular mesangium and often leads to

progressive renal dysfunction and kidney failure. We have previously shown that the mesangial IgA is likely to derive from the bone marrow plasma cells, and suggested that a primary abnormality within the mucosal

immune system may underly the pathogenesis of IgAN. This study has analyzed the T cell receptor (TCR) variable (V) region expression by yfi T cells in the intestinal mucosa of patients with IgAN. The Vy and V usage of TCR transcripts was determined using a semiquantitative reverse transcriptase-PCR protocol. Primers specific for the four human Vy and six V subfamilies were used each with a constant (C) y or C specific primer, and the PCR-amplifled TCR transcripts were detected by Southern blotting and oligonucleotide hybridization. yfi TCR V region expression was determined in gut biopsies and peripheral blood of 11 patients with IgAN, and the TCR Vy and V repertoires were compared to those in gut and peripheral blood of Ii control individuals. yfi T cells in normal blood predominantly expressed Vy2 (Vy9 gene) and V2 gene segments whereas those in normal gut mainly expressed Vy3 and V3. In IgAN patients, V2 was also the predominant V gene utilized by peripheral

blood y T cells, however, we observed a predominance of Vy3 and reduced Vy2 usage by these cells. yfi T cells in the gut of IgAN patients mainly used Vy3 and VM. While the y and TCR V region repertoires did not differ significantly between the peripheral blood of patients and

controls, there were significant differences in Vy and Vfi repertoire expression between IgAN and control gut biopsies. Vy3 gene expression was significantly decreased in IgAN gut compared to control gut (P = 0.023). In addition, there was a significant decrease in Vfi3 gene expression in IgAN gut compared to control gut (P 0.043). These findings indicate

that a subpopulation of y T cells, which represent the majority of y T cells in normal gut mucosa, are significantly diminished in the gut of patients with IgAN. This suggests that a "hole" in the mucosal y T cell repertoire may play a fundamental role in contributing to the pathogenesis of IgAN.

IgA in the glomerular mesangium [3, 4], but the site of synthesis of this IgA, the reasons for its deposition within the mesangium, and the pathogenetic mechanisms leading to glomerular damage remain unclear. It is generally accepted that the underlying abnormality in IgAN

is within the IgA immune system rather than the kidney. IgA deposits reappear in about half of normal kidneys transplanted into recipients with IgAN [5], and, conversely, there have been reports of the disappearance of IgA from kidneys of IgAN donors inadvertently transplanted into non-IgAN recipients [6, 7]. Mesangial IgA in IgAN is known to be of the IgAl subclass, and to be polymeric, containing J-chain [81. Polymeric IgA (pIgA) is nor-

mally produced in large quantities at mucosal surfaces, from where it is transported across the epithelium into mucosal secre-

tions, forming part of the primary defence against potential pathogens. Very little of this pIgA reaches the circulation; serum IgA is mostly monomeric and is derived from the bone marrow. In IgAN, circulating levels of IgAl polymer are raised [8], particu-

larly in response to immunization [9, 10]. These observations, along with the association of clinical disease exacerbations with mucosal infection [11], have led to the assumption that mesangial IgA is derived from the mucosal IgA system in IgAN. However,

we have demonstrated that pIgA production in the duodenal lamina propria is strikingly down-regulated in IgAN [12], while it

is up-regulated in the bone marrow [13—16]. These findings suggest that the bone marrow is the likely source of mesangial IgA

in IgAN, but that the increased bone marrow production may be secondary to a fundamental flaw in mucosal IgA immunity. Since IgA production is T cell dependent, we investigated the possibility

that the failure of pIgA production in the duodenal lamina IgA nephropathy (IgAN) is one of the most common forms of propria in IgAN is secondary to a defect at this site of a specific T glomerulonephritis worldwide and often leads to end-stage renal cell subset, namely y8 T cells. The physiological role of y6 T cells is unclear. While y8 T cells failure [1, 2]. The diagnostic hallmark of IgAN is deposition of share many features of af3 T cells including helper and cytotoxic effector functions, and cytokine production [17], there are fundamental differences between these two cell types, particularly with Key words: T cell, gamma delta T cell, IgA nephropathy, V region, respect to antigen recognition. Unlike a T cells, which require mucosal immune system. MHC molecules for antigen presentation, yfi T cells rarely see antigen in an MHC-restricted manner [17]. In addition, mycobacReceived for publication February 10, 1997 and in revised form May 20, 1997 terial heat shock proteins appear to be preferentially recognized Accepted for publication May 20, 1997 by y T cells [18, 19], although a number of other bacterial ligands © 1997 by the International Society of Nephrology that stimulate yfi T cells have been identified [20—22]. These

1047

1048

Olive et al: y3 T cell receptor expression in IgAN

Table 1. Clinical details of 11 subjects with IgA nephropathy Disease duration

Gender Age M M M M M M M M M

53

F

50 57

36 26 49 53

42 60 34 51

years

Light microscopy

7 10 3 14 4 1 13 24 8 8 8

minimal change minimal change diffuse mesangial hypercellularity diffuse mesangial hypercellularity minimal change segmental hypercellularity mesangial matrix expansion segmental hypercellularity segmental hypercellularity minimal change mesangial matrix

Immune deposits

Past history of macroscopic hematuria

IgA, 1gM IgA only IgA, C3 IgA, C3 IgA, C3 TgA only IgA only IgA only IgA, C3 IgA only IgA, C3

No No Yes Yes No No No Yes No No No

Proteinuria (urine stick test)

Serum creatinine° pmol/liter

negative negative negative negative negative negative

92 95 75

80 97 94

+++

250

negative

124 110

++

Drugs

none none none

anti-hypertensives none none anti-hypertensives anti-hypertensives anti-hypertensivcs none

65 negative 100 none negative Proteinuria was not quantified at the time of study; but in both patients with stick-positive proteinuria serum albumin was normal.

M

include low molecular weight non-peptidic compounds [20], and phosphorylated molecules representing naturally occurring products of bacterial metabolism [21, 22]. y T cells represent a minor population of circulating peripheral blood T cells [23] but have an increased frequency among intra-epithelial lymphocytes (TEL) within the gut mucosa of both the small intestine [24] and colon IEL contribute to host defence [25]. Therefore, it is likely that

symptoms suggestive of gastrointestinal disease and no mucosal

infection or macroscopic hematuria at the time of study; none were nephrotic. None of the patients had ever been treated with corticosteroids or immunosuppressive drugs. Mean serum creatinine was 108 xmol/liter (range 65 to 250 imol/liter). Only one patient had serum creatinine raised above the reference range; this patient subsequently developed progressive renal failure. against environmental antigens that have gained access to the Eleven age- and sex-matched controls (9 male, mean age 43.5

mucosal system. It has been suggested that mucosal immunity may years, range 22 to 74 years) were examined. Controls were be impaired in patients with IgAN, which allows antigens to access individuals undergoing upper gastrointestinal tract evaluation in

the circulation leading to systemic IgA production and the whom gastric and duodenal mucosae appeared macroscopically

deposition of IgA in the glomerular mesangium [8]. In view of this, we have examined IgAN patients for a potential defect in their mucosal yfi T cell repertoire. y T cells express a T cell receptor (TCR) composed of a y and 1 TCR chain [231. The TCR chains are comprised of individual variable (V), diversity, joining, and constant (C) region gene segments that undergo rearrangement during T cell ontogeny to produce functionally rearranged TCR V region genes [23]. In comparison to af3 T cells there are considerably fewer Vy and V genes. In humans, there are eight functional Vy genes in four Vy subfamilies [23]. The V-yl subfamily contains five of these Vy gene segments (Vy2, 3, 4, 5 and 8) whereas subfamilies Vy2, Vy3

and Vy4 each contain a single V gene segment; Vy9, VylO and Vyll, respectively [23]. Six V6 genes have been identified [231. To determine the diversity of y T cell populations in IgAN, this study has analyzed the V region repertoire of yfi TCR in the intestinal mucosa and peripheral blood of patients with IgAN and compared this to they TCR V region repertoire in gut and blood of controls. y TCR V region expression was determined using a semiquantitative reverse transcriptase-PCR employing primers specific for the four human Vy and six V subfamilies, each with a C-y or C specific primer. PCR-amplified TCR transcripts were detected by Southern blotting and oligonucleotide hybridization. The data suggest that a "hole" in the mucosal y5 T cell repertoire may play a fundamental role in contributing to the pathogenesis of

normal and in which subsequent microscopic examination was also normal. No gastrointestinal pathology was identified in any of the controls. Patients and controls were not receiving agents known to interfere with mucosal function. Peripheral blood and gut biopsies Peripheral blood was collected from all individuals at the time when gut biopsies were taken. PBMC were separated following density gradient centrifugation over Ficoll (Pharmacia, North Ryde, NSW, Australia). Endoscopic biopsies from the second part

of the duodenum were obtained from patients and controls. Endoscopies were performed entirely as part of the research protocol for this study which had approval from the appropriate ethics committee. RNA extraction and eDNA synthesis

RNA was extracted from PBMC and gut biopsies using a single-step guanidinium thiocyanate/phenol chloroform procedure according to Chomczynski and Sacchi [26]. RNA yield was estimated by spectrophotometry and denaturing agarose gel dcctrophoresis after staining with ethidium bromide and comparison

IgAN.

with known standards. Approximately 2 p.g total RNA was

METHODS

converted to eDNA by incubation for one hour at 37°C in a 50 1.d reaction containing 50 mM Tris/HC1 (pH 8.3), 75 mivi KC1, 3 mM

Patients and controls Eleven patients (10 male, mean age 46,4 years, range 26 to 60 years) with renal biopsy-proven IgAN were examined. Clinical

details of the patients are shown in Table 1. Patients had no

MgCI2, 10 mvi dithiothreitol, 1 ifiM of each dNTP, 500 ng oligo dT,

25 units RNAsin (Promega, Sydney, Australia) and 50 units M-MuLV reverse transcriptasc (Pharmacia). Reactions were purified by phenol/chloroform extraction.

1049

Olive et al: y T cell receptor expression in IgAN

Semiquantitation of PCR Twofold serial dilutions of a known concentration of pooled PBMC cDNA obtained from four normal healthy individuals were used for the construction of GAPDH and TCR C6 PCR standard curves, as previously described [271. Briefly, PCR amplification was carried out in triplicate for each dilution of eDNA using the forward 5'-ACCACAGTCCATGCCATCAC-3' and reverse 5'TCCACCACCCTGTTGCTGTA-3' GAPDH primers, and the forward CMV 5'-TcAGCCTCATACCAAACCATC-3' and reverse CM 5 '-GACAGCAYT'GTACTflTCCCACTGG-3' TCR C6 primers, respectively. PCR conditions for both GAPDH and TCR C6 were one minute each at 95°C, 60°C and 72°C for 30 cycles. The specificity of each PCR product (5 jd) was confirmed by Southern blotting and hybridization with internal GAPDH (5'-

GATGACATCAAGAAGGTGGTG-3') and TCR CMI (5'TTCACCAGACAAGCGACA-3') oligonucleotide probes that were end-labeled with y32P-ATP. Two dimensional densitometry

was undertaken on the autoradiographs using a computerized

eDNA samples were identical. Amplification of TCR Vy and V6 transcripts was carried out for one minute each at 95°C, 50°C and 72°C for 35 cycles, or 40 cycles for amplification involving Vy4, V64 and V66. PCR products (5 fL!) were electrophoresed in a

1.25% agarose gel and the length of the products assessed by comparison with molecular size standards after staining with ethidium bromide. The specificity of each PCR product was confirmed by Southern blotting and hybridization with internal TCR Cy (5'-GGAAACATCTGCATCAAGTFGT-3') and TCR CMII (5'-GATGGTTTGGTATGAGGCTGA-3') probes that were end-labeled with y12P-ATP.

Southern blotting and oligonucleotide hybridization PCR products (5 fxl) were transferred to a nylon membrane (Amersham, Castle Hill, NSW, Australia) by Southern blotting. Oligonucleotides directed at segments internal to the amplified TCR C6, TCR Ca, GAPDH, TCR Vy and TCR V6 transcripts were end-labeled with y32P-ATP by incubation for 45 minutes at 37°C in a 20 jxl reaction containing 50 mi Tris/HC1 (pH 7.5), 20

densitometer (Molecular Dynamics) and ImageQuant V 3.0 software. Following densitometry, PCR standard curves were con- mM MgCI2, 20 mivi dithiothreitol, 40 jtCi [-y-32P]ATP (Bresatec, Adelaide, SA, Australia) and 10 units T4 polynucleotide kinase structed over a range of eDNA concentrations. (Pharmacia). Hybridization of blots was performed overnight in 5 X SSPE [5 ms'i EDTA, 50 mrvt NaPO4 (pH 7.4), 900 mM NaCI], Standardization of PBMC and gut cDNA samples An aliquot of each eDNA sample was used for PCR amplifica- 5 X Denhardt's solution (0.1% each bovine serum albumin, Ficoll tion of GAPDH and TCR C6 using the same conditions as those 400 and polyvinylpyrrolidone), 0.1% SDS and 100 xg/ml heatused for construction of the PCR standard curves. PCR products denatured calf thymus DNA (Sigma, Castle Hill, NSW, Austrawere Southern blotted and hybridized with internal GAPDH and lia). Membranes were washed twice with 2 X SSC/0.1% SDS for 10 minutes each at room temperature and once with 0.1 X TCR C6 oligonucleotide probes, respectively. Following densi- SSC/0.1% SDS for 10 minutes at 45°C, air-dried and then exposed tometry on each hybridizing DNA band, the concentration of each

PCR amplification was repeated to ensure that amplification

to autoradiographic film. Two dimensional densitometry was undertaken on the autoradiographs using a computerized densitometer (Molecular Dynamics) and ImageQuant V 3.0 software.

remained within the limits of the standard curves.

Statistical and data analysis

cDNA sample was estimated by extrapolation from the PCR standard curves. eDNA samples were diluted accordingly and the

PCR amplification of TCR Cl and TCR Ca transcripts

The TCR C6 primers, CMV and CM, were used for PCR amplification of TCR C6 transcripts in the cDNA samples. PCR-amplified TCR C8 transcripts were identified by hybridiza-

tion with the CMI probe. For PCR amplification of TCR Ca transcripts in the cDNA samples the forward 5'-AATATCCAGAACCCTGACCCT-3' and reverse 5-GCTCCAGGCCACAG-

CACTGTTGCT-3' TCR Ca primers were used with the same conditions as those used for TCR 6 chain amplification. PCRamplified TCR Ca transcripts were detected by hybridization with

an internal Ca probe (5'-CAGAC1TGTCACTGGA-3').

PCR amplification of TCR Vy and V chain transcripts cDNA samples standardised for TCR C6 gene expression were subjected to V region specific PCR amplification of rearranged TCR V-C transcripts. Each 50 d PCR reaction contained 10 mivi Tris/HCI (pH 8.4), 50 ms KCI, 1.5 mM MgCl2, 0.01% (wt/vol) gelatin, 0.2 m'vi of each dNTP, 25 pmole each of forward (V region) and reverse (C region) primer, and 1.25 units Taq DNA polymerase (Promega). Primers for the four TCR Vy and six V6 subfamilies, and those for Cy and C6 have previously been published [28]. Pooled normal PBMC eDNA was used as a positive control for PCR amplification. Negative PCR controls contained all the reaction components for PCR except for eDNA template. PCR conditions for amplification of PBMC and gut

To account for variability in the amount of PBMC or gut sample, RNA recovery and efficiency of reverse transcription, TCR C6 and TCR Ca gene expression were determined as a ratio of GAPDH gene expression. TCR Vy and V6 subfamily expression were determined as a ratio of the TCR C6 densitometry score

after each sample had been standardized for TCR C6 gene expression. Control peripheral blood and control gut group scores

were compared to those of the IgAN group using a MannWhitney U-test. Paired samples were analyzed using the Wilcoxon

rank sum analysis. P < 0.05 was accepted as statistically significant. RESULTS

RNA samples From the PBMC samples, 20 RNA samples were obtained (10

patient; 10 control) that were suitable for subsequent TCR analysis. Of the gut biopsies, 11 control and 9 patient RNA samples were obtained. Those samples with undetectable or degraded RNA were discarded.

PCR standard curves and standardization of eDNA samples PCR standard curves were constructed for GAPDH and TCR C6 using a range of eDNA concentrations obtained from pooled normal PBMC (Fig. 1). A logistic regression analysis of the curves

showed that for GAPDH r2 =

0.87

and for TCR C6 r2 =

0.92.

1050

Olive et al: 'yS T cell receptor expression in I4AN

A

35

2200 2000

'a) og
f

1400 1200

.2 E

1000

U)

0)

*

30 25

20

**

I

I

f

4

15

I

f

10

0 600

5

400

200

4

a

0

0 Oct

0 0.005

0.025

0.5

0.125

CS

Fig. 2. TCR Ca and TCR CS gene expression in peripheral blood and gut

biopsies of IgAN patients and controls. The Ca and CS signals were expressed as a ratio to GAPDH. Results are plotted as mean standard error. Symbols are: (S) control PBMC; (0) IgAN PBMC; () control gut; (LII) IgAN gut; P = 0.005; ** = 0.008.

cDNA concentration B 1600 1400

0 0 Ci)

1200

f

1000

of

140 120 a)

0 0

800

0

100

(1)

>E

600

o 400

80 60

U)

a)

C a)

200

20

0

0

0.005

40

0.025

0.125

0.5

cDNA concentration Fig. 1. PCR standard curves for GAPDH (A) and TCR CS (B). Twofold serial dilutions of pooled normal PBMC cDNA were subjected each in triplicate to PCR amplification of GAPDH and TCR CS. PCR products were detected by Southern blotting and hybridization with an internal oligonucleotide probe. Following densitometry, the signals were plotted as mean standard error.

0 V'1 Vy4 Vy2 V''3 Fig. 3. TCR V7 repertoire expression in peripheral blood and gut biopsies of IgAN patients and controls. Vy signals were expressed as a ratio to TCR C& Results are plotted as mean standard error. Symbols are: (•) control PBMC; (0) IgAN PBMC; (•) control gut; (LI) IgAN gut; = 0.017; *** = 0.021; **** = 0.023. *p = 0.047;

peripheral blood of IgAN patients and controls, yet there was a significantly lower level of Ca gene expression in the gut samples Following PCR amplification of GAPDH and TCR CS in the of both patients and controls (patient PBMC vs. gut, P = 0.008; blood and gut cDNA samples, the concentration of each cDNA control PBMC vs. gut, P = 0.005). In contrast, CS gene expression sample was estimated by comparison of the densitometry signal was relatively high in the PBMC and gut cDNA samples of obtained with the PCR standard curves. cDNA samples were patients and controls with no significant differences observed diluted accordingly and PCR amplification of GAPDH and TCR between the two groups. CS was repeated to confirm that the signals remained within the TCR Vy and VS repertoire expression in control blood and gut limits of the standard curves. The usage of Vy and VS genes by yS T cells in the peripheral TCR aj3 and TCR yS expression in controls and IgAN patients blood of controls showed the predominance of Vy2 (Vy9) and Expression of the af3 TCR and yS TCR (Fig. 2) was determined V52 gene segments (Figs. 3 and 4). In addition to Vy2, TCR in PBMC and gut cDNA samples by PCR amplification using transcripts using either Vyl or Vy3 were also readily detected in primers specific for the Ca and CS region of the TCR, respec- the peripheral blood whereas Vy4 expression was low in compartively. To correct for the amount of tissue available in each ison (Fig. 3). V52 was predominantly expressed by normal periphsample, results were expressed as a ratio to GAPDH gene eral yS T cells, although all of the other VS genes were detected expression. Ca gene expression was readily detected in the with VS5 being the second most utilized VS gene (Fig. 4). In the

Olive et al: y6 T cell receptor expression in IgAN

DISCUSSION

120 a)

0

1051

Human TEL in the gastrointestinal tract most likely play a

100

crucial role in mucosal immunity [29]. While the majority of these

C-)

(I)

80

IEL bear af3 TCR [29], the expression of the y6 TCR by IEL is also prominent [24, 25]. This study found that TCR c43 and TCR

>E

60

ci)

40

y6 expression were relatively high in the peripheral blood of TgAN patients and controls. However, TCR a/3 expression was significantly decreased in the gut biopsies of both patients and controls, whereas TCR y6 expression was high in all gut biopsies with no

C a)

20 0 V61

V62

V63

V64

V5 V66

Fig. 4. TCR V6 repertoire expression in peripheral blood and gut biopsies of IgAN patients and controls. V6 subfamily signals were expressed as a ratio to TCR C6. Results are plotted as mean standard error. Symbols are: (•) control PBMC; (0) IgAN PBMC; (U) control gut; (0) IgAN gut; * 0.017; ** = 0.043; ***p = 0.049.

significant differences observed between the peripheral and gut compartments of either group. These findings indicate that a significant proportion of the T cells in duodenal mucosa, whether derived from patients with IgAN or control individuals, were y6T cells.

This study has characterized the populations of y6 T cells in gut mucosa of patients with IgAN. A semiquantitative reverse transcriptase-PCR protocol was used to determine the Vy and V6 repertoire expression by y6 T cells in gut biopsies and peripheral blood of a group of IgAN patients and controls. Several studies have previously analyzed the expression of the y6 TCR by TEL

residing in the normal intestinal mucosa using V61 and V62 control gut biopsies, TCR y transcripts mainly used Vy3 although TCR transcripts utilizing Vyl, Vy2 and Vy4 were also detected

(Fig. 3). TCR 6 chain expression in control gut was restricted

region specific monoclonal antibodies [25, 30, 311, and by PCR analysis [31, 32]. In contrast to normal peripheral blood in which the majority of y6 T cells expressed a receptor encoded by Vy9

largely to V63, and with the exception of V66, all of the other V6 genes were detected but to a lesser degree (Fig. 4). By comparison

and

of the control blood and gut group scores, we found that the normal peripheral and gut y6 TCR V region repertoires were significantly different with respect to TCR y chain expression. Expression of Vyl was significantly decreased in the gut com-

quantitative PCR analysis of TCR transcripts that normal y6 T cells in the periphery preferentially utilized Vy9 and V62. In addition, our data demonstrate that the Vyl subfamily is the next most frequently used Vy subfamily in normal peripheral blood y6 T cells, followed by Vy3 and Vy4. This finding supports that of a recent report that analyzed the repertoire of Vy genes expressed

pared to blood (P = 0.047). Similarly, reduced expression of Vy2 was observed in gut compared to blood (P = 0.017). TCR Vy and V& repertoire expression in IgAN blood and gut

In IgAN patients, we observed a predominance of Vy3 and

V62 TCR chains [33], those in the gut predominantly

expressed V81 TCR [25, 30—321. We have also shown by semi-

by normal peripheral blood y6 T cells using Vy specific monoclo-

nal antibodies [31. Our PCR strategy, represents, therefore, an accurate measure of y6 TCR expression levels. In normal gut mucosa, we observed a predominance of V63, as

reduced Vy2 usage by y6 T cells in peripheral blood (Fig. 3). Vyl and Vy4 transcripts were also detected. As with control PBMC, opposed to V61 [25, 30—32]. Other investigators have also shown, V62 was the predominant V6 gene utilized by peripheral blood y6 by PCR analysis, that V63 gene expression was prominent among T cells in IgAN and again all of the other V6 genes were detected mucosal y6 TEL [32, 35]. In previous studies of y6 TCR expression with the exception of V64 (Fig. 4). y6 T cells in the gut of IgAN in normal gut, however, V63 gene expression was not analyzed patients mainly used Vy3 and V61 (Figs. 3 and 4). TCR 6 chain due to the absence of a V63 specific monoclonal antibody, which expression in IgAN gut, however, was remarkably low for all six may explain the discrepancy. By using V region-specific PCR V6 genes (Fig. 4). By comparison of the peripheral blood and gut amplification of TCR transcripts, we were able to identi!' the group scores for 1gAN patients, we found that three V region other V6 genes used by TCR 6 chain transcripts in normal gut.

genes each had a significantly decreased expression in the gut Expression of V61, V&2, V64 and V65 genes was observed in compared to blood (Vy2, P = 0.021; V62, P = 0.017; V65, P = mucosal 6 chain trancripts, but to a lesser degree, whereas 0.049).

Comparison of the 'y& TCR V region expression between IgAN

patients and controls While the y and 6 TCR V region repertoires did not differ significantly between the peripheral blood of IgAN patients and controls, we observed significant differences in Vy and V6 repertoire expression between IgAN and control gut biopsies. Vy3 gene expression was significantly decreased in TgAN gut compared to

expression of V66 was not detected. Bucht et al [31] also observed significant expression of V62 and V65 by TCR 6 chain transcripts

in the colon. There was a predominant expression of Vy3 in normal gut although TCR transcripts utilizing Vyl, Vy2 and Vy4 were detected. The usage of individual Vy genes by yö T cells in normal duodenal mucosal biopsies has previously been shown to be biased towards Vyl and Vy3 [35]. Overall, we found significant

differences in the normal peripheral and gut y6 TCR V region

In addition, there was a significant

repertoires with respect to TCR y chain expression; expression of VyT and Vy2 was significantly decreased in the gut compared to

decrease in V63 gene expression in IgAN gut compared to control gut (P = 0.043). Of note, Vy3 and V63 were the predominant V region genes expressed by y8 T cells in control gut biopsies.

blood. These differences may reflect the different antigens encountered by y6 T cells in the periphery and small bowel mucosa. In the peripheral blood of IgAN patients, Vy3 and V62 were

control gut (P =

0.023).

1052

Olive et al: yb T cell receptor expression in IgAN

the predominant V region genes utilized by yb T cells. Those in the gut of IgAN patients mainly used Vy3 and Vbl. By compar-

mality. Coeliac disease occurs with an increased frequency in

patients, we found that Vy2, Vb2, and Vb5 each had a significantly decreased expression in the gut compared to blood. Furthermore,

mechanisms leading to a defect in the mucosal yb T cell repertoire

ison of the peripheral blood and gut group scores for IgAN

while the y and b TCR V region repertoires did not differ

patients with IgAN [43], further supporting an underlying primary

abnormality within the mucosal immune system in IgAN. The in this disease, however, are speculative. In mice the intestinal epithelium is an extrathymic location where a highly restricted subset of y T cells are generated [44]. Although no extrathymic

significantly between the peripheral blood of IgAN patients and controls, y TCR V region expression was significantly different in site for yfi T cell programmed differentiation has yet been IgAN gut compared to control gut biopsies. Vy3 and Vb3 gene demonstrated in the human, intense speculation on the origins of expression were significantly decreased in IgAN gut compared to IEL is ongoing. The alteration in yfi T cell repertoire in patients control gut, whereas Vy3 and Vb3 were the predominant V region with IgAN may arise from altered processing of yb T cells during genes expressed by yfi TCR in control gut biopsies. These findings extrathymic ontogeny, or in response to antigen [45] or hormonal indicate that a subpopulation of yb T cells which represent the stimuli [46]. majority of yfi T cells in normal gut mucosa are significantly In summary, our findings suggest that a defect in the mucosal yfi diminished in the gut of patients with IgAN, suggesting that there T cell repertoire contributes to an abnormality within the mucosal is a "hole" in the mucosal y T cell repertoire in this disease. immune system, and the pathogenesis of IgAN. We have previously demonstrated in IgAN a reduction in the proportion of lamina propria IgA plasma cells expressing J chain ACKNOWLEDGMENTS mRNA, and by inference, therefore, a reduction in pIgA producWe thank the Wellcome Trust for their support of Dr. Steven Harper tion [121. The production of J chain, which links Ig molecules to (Grant 03493/91). This work was supported in part by the Australian form polymers, is T cell dependent [361. The present demonstra- Kidney Foundation, and the Renal Research Trust Fund (Princess tion of a "hole" in the mucosal yb T cell repertoire underpins Alexandra Hospital).

these previous findings and lends credence to the view that a fundamental flaw exists in the mucosal IgA immune system in IgAN.

Recent experiments indicate that mucosal immunity may be

Reprint requests to Dr. Colleen Olive, Department of Renal Medicine, University of Queensland, Princess Alexandra Hospital, Ipswich Road, Woolloongabba, Brisbane, Queensland 4102, Australia. E-mail: [email protected]

impaired in yb T cell deficient mice [37]. These mice had

decreased IgA plasma cells in their mucosa, and when adminis- REFERENCES 1. D'Arviico G: The commonest glomerulonephritis in the world: IgA tered antigen orally were shown to have reduced IgA antibody nephropathy. Quart J Med 245:709—727, 1987 production compared to control mice [37]. If human yb T cells 2. EMANCIPATOR SN, LAMM ME: IgA nephropathy: Pathogenesis of the also mediate mucosal IgA responses, it is possible that a "hole" in most common form of glomerulonephritis. Lab Invest 60:138—168, the repertoire of yfi T cells, as we observed in the mucosa of IgAN

patients, may lead to fewer IgA plasma cells in the gut, and

1989 3. BERGER J: IgA glomerular deposits in renal disease. Transplant Proc

impaired IgA mucosal immunity to specific antigen. In a recent study of immune responses to nasally-administered

4. GALLA JH: Perspectives in clinical nephrology: IgA nephropathy.

1:939—944, 1969

nasal washings of patients with IgAN, in contrast to controls [381.

Kidney mt 47:377—387, 1995 5. ODUM J, PEH CA, CLARKSON AR, BANNISTER KM, SEYMOUR AE, GILLIS D, THOMAS AC, MATFHEW TH, WOODROFFE AJ: Recurrent

This finding suggests there is indeed a failure of mucosal IgA

mesangial IgA nephritis following renal transplantation. Nephrol Dial

immunity in IgAN. Increased production of pIgA in the bone marrow and increased circulating pIgA levels may then be sec-

Transplant 9:309—312, 1994 6. SANFILIPPO F, CROKE BP, BOLLINGER R: Fate of four cadaveric renal

ondary to this decreased mucosal production. In this context, it is of interest to note that, while we did not observe any alteration in

1982 7. SILVA FG, CHANDER P, PIRANI CL, HARDY MA: Disappearance of

the V region usage of yb T cells in the blood of patients with IgAN, it has been reported that yfi T cells account for a higher percentage of circulating CD3 T cells in IgAN, and that this

glomerular mesangial IgA deposits after renal allograft transplanta-

cholera toxin, no antigen-specific IgA could be detected in the

difference becomes more marked after immunization with tetanus toxoid [39]. We have also demonstrated increased numbers of yb T cells infiltrating the renal interstitium in IgAN, but only in those

patients with progressive renal impairment [40], though it is probable that this finding identifies a role for yb T cells in the "downstream" events of progressive renal disease rather than the primary mechanisms underlying mesangial IgA deposition.

yb T cells have also been implicated in the pathogenesis of

coeliac disease, in which gluten-intolerance leads to lesions in the

small intestinal mucosa involving a T cell-mediated immune response [41, 42]. There is evidence that y T cells are increased in the small bowel mucosa of patients with coeliac disease [30]. Moreover, Falk et al [35] have shown that the y TCR V region repertoire in coeliac disease gut differs compared to control gut, which suggests that y T cells contribute to the mucosal abnor-

allografts with mesangial IgA deposits. Transplantation 33:370—376,

tion. Transplantation 33:214—216, 1982

8. FEEHALLY J: Immune mechanisms in glomerular IgA deposition. Nephrol Dial Transplant 3:361—378, 1988 9. VAN DEN WALL BAKE AWL, BEYER WEP, EVERS-SCHOUTEN JH, HERMANS J, DAHA MR, MASUREL N, VAN Es LA: Humoral immune

response to influenza vaccination in patients with primary immunoglobulin A nephropathy. J Clin Invest 84:1070—1075, 1989 10. LAYWARD L, ALLEN AC, HARPER SJ, HATFERSLEY JM, FEEHALLY J:

Increased and prolonged production of specific polymeric IgA after systemic immunisation with tetanus toxoid in IgA nephropathy. C/in Exp Immunol 88:394—398, 1992 11. EMANCIPATOR SN: Immunoregulatory factors in the pathogenesis of IgA nephropathy. Kidney mt 38:1216—1229, 1990 12. HARPER SJ, PRINGLE JH, WICKS ACB, HAVrERSLEY J, LAYWARD L, ALLEN A, GILLIES A, LAUDER I, FEEHALLY J: Expression of J chain

mRNA in duodenal IgA plasma cells in IgA nephropathy, Kidney Int 45:836—834, 1994 t3. VAN DEN WALL BAKE AWL, DAHA MR, RADL J, HAAIJMAN JJ, VAN

DEN ARK A, VALENTIJN RM, VAN Es LA: The bone marrow as production site of the tgA deposited in the kidneys of patients with IgA nephropathy. Cl/n Exp Immunol 72:321—325, 1988

Olive et al: y T cell receptor expression in IgAN 14. VAN DEN WALL BAKE AWL, DAHA MR, HAAIJMAN JJ, RADL J, VAN

DEN ARK A, VAN Es LA: Elevated production of polymeric and monomeric IgAl by the bone marrow in IgA nephropathy. Kidney mt 35:1400—1404, 1989 15. HARPER SJ, ALLEN AC, LAYWARD L, HATTERSLEY H, VEITCH PS,

FEEHALLY J: Increased immunoglobulin A and immunoglobulin Al cells in bone marrow trephine biopsy specimens in immunoglobulin A nephropathy. Am J Kid Dis 24:888—892, 1994 16. HARPER SJ, ALLEN AC, PRINGLE JH, FEEHALLY J: Increased dimeric

IgA producing B cells in the bone marrow in IgA nephropathy determined by in situ hybridisation for J chain mRNA. J Clin Pathol 49:38—42, 1996 17. HAAS W, PEREIRA P, TONEGAWA S: 11:637—685, 1993

y/ cells. Annu Rev Immunol

18. HAREGEWOIN A, SOMAN G, HOM RC, FINBERG RW: Human yh T

1053

associated gastritis, coeliac disease and inflammatory bowel disease. C/in Exp Immunol 84:440—444, 1991 31. BUCHT A, SODERSTROM K, ESIN 5, GRUNEWALD J, HAGELBERG 5, MAGNUSSON I, WIGZELL H, GRONBERG A, KIESSLING R: Analysis of

yb V region usage in normal and diseases human intestinal biopsies and peripheral blood by polymerase chain reaction (PCR) and flow cytometry. C/in Exp Immunol 99:57—64, 1995 32. HOLTMEIER W, CHOWERS Y, LUMENG A, MORZYCKA-WROBLEWSKA

E, KAGNOFF MF: The T cell receptor repertoire in human colon and peripheral blood is oligoclonal irrespective of V region usage. J Clin Invest 96:1108—1117, 1995 33. TRIEBEL F, FAURE F, MAMI-CHOUAIB F, JITSUKAWA S, GRISCELLI A, GENE VEE C, ROMAN-ROMAN 5, HERCEND T: A novel human Vb gene

expressed predominantly in the TiyA fraction of y/b peripheral lymphocytes. Eur J Immunol 18:2021—2027, 1988

cells respond to mycobacterial heat shock protein. Nature 340:309—

34. SCHONDELMAIER S, WESCH D, PECHHOLD K, KABELITZ D: Vy gene

312, 1989 19. BORN W, HAPP MP, DALLAS A, REARDON C, KUB0 R, SHINNICK T,

35. FALK MC, NG G, ZMANG GY, FANNING GC, KAMATI-I KR, KNIGHT JF:

BRENNAN P, O'BRIEN R: Recognition of heat shock proteins and y cell function. Immunol Today 11:40—43, 1990

usage in peripheral blood yb T cells. Immunol Len 38:121—126, 1993

Predominance of T-cell receptor Vb3 in small bowel biopsies from

K, WAGNER H: A lectin-binding, protease-resistant mycobacterial

coeliac disease patients. Clin Exp Immunol 98:78—82, 1994 36. BREwER JW, RANDALL TD, PARKHOUSE RME, CORLEY RB: 1gM hexaniers? Immunol Today 15:165—168, 1994

148:575—583, 1992

37. FUJIHASHI K, MCGHEE JR, KSEON MN, COOPER MD, TONEGAWA 5, TAKAI-IASHI I, HIROI T, MESTECKY J, KIYONO H: y/b T cell-deficient

20. PFEFFER K, SCHOEL B, PLESNILA N, LIPFORD GB, KROMER S, DEUSCH

ligand specifically activates Vy9 human yh T cells. J Immunol 21. CONSTANT P, DAVODEAU F, PEYRET M-A, POQUET Y, Puzo G, BONNEVILLE M, FOURNIE J-J: Stimulation of human y T cells by nonpeptidic mycobacterial ligands. Science 264:267—270, 1994

22. BURK MR, M0RI L, DE LIBERO G: Human Vy9-Vh2 cells are

mice have impaired mucosal immunoglobulin A responses. J Exp Med 183:1929—1935, 1996 38. DE FIJTER JW, ELIGENRAAM JW, BRAAM CA, HOLMGREN J, DAnA MR, VAN Es LA, VAN DEN WALL BA AWL: Deficient IgAl immune

stimulated in a cross-reactive fashion by a variety of phosphorylated metabolites. Eur J immunol 25:2052—2058, 1995 23. RAIJ LET D: The structure, function, and molecular genetics of the yIiS T cell receptor. Annu Rev Immunol 7:175—207, 1989

Kidney mt 50:952—996, 1996 39. FORTUNE F, COURTEAU M, WILLIAMS DG, LEHNER T: T and B cell

24. JARRY A, CERF-BEN5USSAN N, BROIJSSE N, SELZ F, GUY-GRAND D: Subsets of CD3 (T cell receptor o/j3 or ylh) and CD3-lymphocytes

40. FALK MC, NG G, ZHANG GY, FANNING GC, PAUL ROY L, BANNISTER

isolated from normal human gut epithelium display phenotypical features different from their counterparts in peripheral blood. Eur

Jlmmunol 20:1097—1103, 1990 25. DEUSCH K, LULING F, REICH K, CLASSEN M, WAGNER H, PFEFFER K:

A major fraction of human intraepithelial lymphocytes simultaneously

expresses they/5 T cell receptor, the CD8 accessory molecule and preferentially uses the Vfil gene segment. Eur J Immunol 21:1053—

1059, 1991 26. CI-IOMCZYNSKI P, SACCHI N: Single-step method of RNA isolation by

acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 162:156—159, 1987

27. OLIVE C, NIcOL 0, FALK MC: Characterisation of y5 T cells in renal cell carcinoma patients by PCR analysis of T cell receptor transcripts. Cancer Immunol Immunother 44:27—34, 1997 28. OLIVE C, GATENBY PA, SERJEANTSON SW: Variable gene usage of T

cell receptor y- and S-chain transcripts expressed in synovia and peripheral blood of patients with rheumatoid arthritis. C/in Exp Immunol 87:172—177, 1992

response to nasal cholera toxin subunit B in primary IgA nephropathy. responses following immunisation with tetanus toxoid in IgA nephropathy. Clin Exp Immunol 88:62—67, 1992 KM, THOMAS AC, CLARKSON AR, WOODROFFE AJ, KNIGHT JF:

Infiltration of the kidney by a/3 and yb T cells: Effect on progression in IgA nephropathy. Kidney mt 47:177—185, 1995 41. MARSH MN: Gluten, major histocompatibihity complex, and the small intestine. A molecular and immunobiological approach to the spectrum of gluten sensitivity ('celiac sprue'). Gastroenterology 102:330— 354, 1992 42. MACDONALD TT, SPENCER J: Evidence that activated mucosal T cells

play a role in the pathogenesis of enteropathy in human small intestine. J Exp Med 167:1341—1349, 1988 43. PASTERNACK A, COLLIN P, MUSTONENE J, REUNALA T, RANTALA I,

LAURILA K, TEPPO AM: Glomerular IgA deposits in patients with cehiac disease. Clin Nephrol 34:56—60, 1990 44. BANDEIRA A, ITOHARA 5, BONNEVILLE M, BURLEN DEFRANOUX 0, MOTA SANTOS T, COUTINI-IO A, TONEGAWA 5: Extrathymic origin of

intestinal intraepithelial lymphocytes bearing T-cell antigen receptor

yb. Proc NatlAcad Sci USA 88:43—47, 1991 45. IMAOKA A, MATSUMOT S, SETOYAMA H, OKADA Y, UMESAKI Y:

29. TREJDOSIEWICZ LK: What is the role of human intestinal intraepithehal lymphocytes? C/in Exp Immunol 94:395—397, 1993 30. TREJD0SIEwIc7 LK, CALABRESE A, SMART Ci, OAKES Di, H0wDLE

Proliferative recruitment of intestinal intraepithelial lymphocytes after microbial colonization of germ-free mice. EurJ Immunol 26:945--

PD, CRABTREE JE, LOSOWSKY MS, LANCASTER F, B0YLST0N W: yT

46. BOLL G, REIMANN J: Oestrogen treatment depletes extrathymic T cells from intestinal lymphoid tissues. Scand J Immunol 43:345—350,

cell receptor-positive cells of the human gastrointestinal mucosa: Occurrence and V region gene expression in Ilelicobacter pylod-

948, 1996 1996