!C Chniicall and
`Expedrinentall
`ilmmun o 11
`
`Editor (cid:9)
`
`R A Thompson
`
`Associate Editor P J Lachmann
`
`tit014 (cid:9)
`c$TRIA__ (cid:9)
`
`t (cid:9)
`q344,, (cid:9)
`
`011?
`
`ritt
`Olt;
`WO: I q7.7.14zil 0.1z 1-4allin,
`
`Ex. 1053 - Page 1
`
`

`

`CLINICAL AND EXPERIMENTAL IMMUNOLOGY
`EDITOR Dr R. A. Thompson
`ASSOCIATE EDITOR Prof. P. J. Lachmann
`
`Clin. ea
`
`G. L. Asherson Harrow
`J. F. Bach Paris
`R. W. Baldwin Nottingham
`A Basten Sydney
`J. R. Batchelor London
`A. G. Bird Edinburgh
`G. F. Bottazzo London
`D. L. Brown Cambridge
`D. Catovsky London
`R. L. Dawkins Perth
`A. M. Denman Harrow
`P. W. Ewan Cambridge
`M. Feldmann London
`A. Ferguson Edinburgh
`M. R. Haeney Salford
`
`EDITORIAL BOARD
`D. B. Lowrie London
`A. Hamblin London
`P. M. Lydyard London
`L. Hammerstrom Huddinge
`R. N. Maini London
`A. R. Hayward Denver
`I. McConnell Edinburgh
`P. Henson Denver
`J. Mowbray London
`J. Ivanyi London
`J. Newsom-Davis Oxford
`G. Janossy London
`G. S. Panayi London
`R. Jefferis Birmingham
`M. Pepys London
`M. Jondal Sweden
`A. J. Pinching London
`J. R. Kalden Erlangen
`L. Poul ter London
`M. Kazatchkine Paris
`J. H. L. Playfair London
`T. Kishimoto Osaka
`M. Reichlin Oklahoma
`S. Knight Harrow
`G. A. W. Rook London
`T. Lehner London
`F. S. Rosen Boston
`F. Y. Liew Beckenham
`M. Steward London
`M. Lockwood Cambridge
`
`W. H. Stimson Glasgow
`D. I. Stott Glasgow
`G. A. T. Targett London
`J. L. Turk London
`R. van Furth Leiden
`T. A. Waldmann Bethesda
`M. J. Walport London
`A. D. B. Webster Harrow
`K. Whaley Glasgow
`D. J. White Cambridge
`P. C. Wilkinson Glasgow
`R. Williams London
`
`T. Fox Desk Editor
`
`Clinical and Experimental Immunology is published monthly.
`Three issues form one volume. The subscription for the 1990
`volumes is £265.50 (UK), £318.00 (overseas), $528.00 (North
`America & Canada). Add to the cost of the regular subscription
`£3.30 or $5.50 per issue for air mail and £1.65 or $2.50 per issue
`for registered post. The Journal is dispatched within Europe by
`surface mail, to other continents by various forms of air speeded
`delivery: to the US* by air freight for forwarding by second class
`post, to India by air freight for guaranteed local delivery, and to
`all other countries by Accelerated Surface Post.
`The Journal is available as a personal subscription to
`members of the British Society for Immunology at a special rate
`through the Treasurer of the Society, and to members of other
`Immunology Societies by application to the publishers.
`Papers describing original research on the role of immuno-
`logy in disease are welcome. The brief will be interpreted widely
`to include case reports of unusual patients with immunological
`disorders, the characterization of new, or the clarification of
`existing disease entities, the use of immunological investigations
`in clinical management, new forms of clinical therapy involving
`immune manipulation, important new clinical diagnostic meth-
`ods, and studies which throw light on the immunopathogenesis
`of disease. These may include experimental studies of animal
`disease and relevant experimental forms of immunotherapy.
`Since a better understanding of normal immunity is a necessary
`prelude to understanding immunopathology, research which
`
`*Second class postage paid at Rahway, NJ. Post Master, send
`address changes to Clinical and Experimental Immunology, c/o
`Mercury Airfreight International, 2323 Randolph Avenue, Avenel,
`NJ 07001, USA.
`
`throws light on the normal human immune system will also he
`welcome.
`Papers considered to report important new advances will he
`given accelerated publication. While authors may request this
`when submitting their manuscripts, this facility is at the
`complete discretion of the Editors.
`
`Editorial correspondence should be directed to the Editor,
`Dr R. A. Thompson, Clinical and Experimental Immunolog)
`East Birmingham Hospital, P.O. Box 1696, BordesleY Green.
`Birmingham B9 5PZ (Tel. 021 766-8125; FAX 021 771-2783).
`
`Business matters, including advertising, remittances to subsoip
`tions, back numbers and offprints, should be sent to the
`publishers Blackwell Scientific Publications Ltd, Osney Mead,
`Oxford OX2 OEL (Tel. 0865 240201; FAX 0865 721205).
`ASCA,
`
`Clinical and Experimental Immunology is covered in
`Current Contents, Index Medicus, ISI/BIOMED and Science
`Citation Index.
`
`© 1990 Blackwell Scientific Publications Ltd. Authorization In
`photocopy items for internal or personal use, or the internal or
`personal use of specific clients, is granted by Blackwell Scientific
`Publications Ltd for libraries and other users registered with*
`Copyright Clearance Center (CCC) Transactional Reporting
`Service, provided that the base fee of $05.00 per copy is paid
`directly to CCC, 27 Congress Street, Salem, Massachusetts
`01970, USA.
`
`0009-9104/90 $05.00
`
`A re&
`lymph
`and cl
`circula
`with d
`of CL
`immu
`effects
`azidot
`severe
`severr
`associ
`cell 11
`
`Cc
`gie, II
`d'Aut
`
`Ex. 1053 - Page 2
`
`

`

`Glasgow
`;ow
`t London
`'on
`rides
`n Bethesda
`.,ondon
`ir Harrow
`pow
`abridge
`Glasgow
`!don
`
`'tor
`
`vill also be
`
`ces will be
`;quest this
`is at the
`
`le Editor,
`imunology
`ey Green,
`71-2783).
`
`subscrip-
`M to the
`ey Mead,
`05).
`
`n ASCA,
`d Science
`
`zation to
`[Mid or
`Scientific
`I with the
`eporring
`y is paid
`tchuserrs
`
`Clin. exp. Immunol. (1990) 81, 177-182
`
`T activation marker evaluation in ARC patients treated with AZT.
`Comparison with CD4+ lymphocyte count in non-progressors and
`progressors towards AIDS
`
`M. LEVACHER, S. TALLET, M. C. DAZZA, E. DOURNON, B. ROUVEIX & J.J. POCIDALO
`Pharmacology Laboratory, INSERM U13, Hdpital Claude Bernard, Paris, France
`
`(Accepted for publication 13 February 1990)
`
`SUMMARY
`
`Reductions in the percentage and absolute number of CD4+ lymphocytes, as well as abnormally high
`levels of activated peripheral T lymphocytes (CD3+ HLA-DR+ phenotype) and an increased
`proportion of CD8+ cells coexpressing the CD57 surface antigen (involved in natural killer activity)
`have been reported in HIV infection and associated with disease progression. We prospectively
`measured these subsets of lymphocytes in 34 patients with advanced AIDS-related complex (ARC)
`treated with azidothymidine (AZT). Peripheral blood lymphocyte phenotyping was performed
`before treatment, then at weeks 12 and 24. A striking fall in the proportion of activated T
`lymphocytes from baseline was observed (P < 0.001) at week 24. In contrast, the percentage of CD4+
`cells showed a slight and transient rise at week 12 (P <0.05). No modification in levels of CD8+ or
`CD8+ CD57+ cells was detected during the study. Of the 34 patients, 11 developed AIDS, and 23
`remained AIDS-free during 51 weeks of follow-up. Similar patterns of change in CD4+ and HLA-
`DR + CD3+ lymphocytes were found in the AIDS progressors and nonprogressors. Likewise, HIV
`p24 antigenaemia showed parallel decreases in both groups of patients. Although changes in CD4+
`cells, p24 antigenaemia and HLA-DR-reactive T lymphocytes were not predictive of clinical
`outcome, large differences existed between the two groups prior to the initiation of therapy. The
`short-term onset of AIDS was associated with lower CD4+ cell numbers, higher levels of serum p24
`antigen and a greater proportion of activated T lymphocytes. Our results suggest that the possible
`interest of T lymphocyte activation markers, in conjunction with conventional phenotyping, should
`be investigated further.
`
`Keywords CD4 lymphocytes activated T lymphocytes ARC AZT
`
`INTRODUCTION
`
`A reduction in the percentage and absolute number of T helper
`lymphocytes is the hallmark of HIV infection. Epidemiologic
`and clinical studies have indicated that the level of CD4+
`circulating lymphocytes is one of the most important correlates
`with disease progression (Moss et al., 1988). Serial measurement
`of CD4+ cells is thus commonly used to detect changes in
`immune status with time, due to disease progression or the
`effects of therapeutic intervention. The clinical benefit of
`azidothymidine (AZT) treatment in patients with AIDS and
`severe AIDS-related complex (ARC) has been reported in
`several trials (Fischl et al., 1987; Dournon et al., 1988) but is
`associated only with a modest and transient increase in CD4+
`cell levels. However, this biological parameter shows large
`Co
`rrespondence: Dr Maryse Levacher, Laboratoire de Pharmacolo-
`gie,
`, INSERM U13, Hopital Claude Bernard, 10, Avenue de la Porte
`d Aubervilliers, 75944 Paris Cedex 19, France.
`
`individual variability (Taylor et al., 1989), making accurate
`assessment difficult. Several other substantial lymphocyte sub-
`set alterations have been reported in HIV-infected individuals
`(Lane & Fauci, 1985). For instance, high levels of peripheral T
`cells expressing antigenic determinants related to a state of
`activation have been described and forwarded as an indicator of
`disease stage (Prince, Arens & Kleinman, 1987; Giorgi & Detels,
`1989). Likewise, impairments of phenotypic subsets associated
`with natural killer (NK) activity suggest that these cells are
`involved in the pathogenesis of HIV infection (Hahn et al.,
`1989).
`We therefore performed a prospective study of the numeri-
`cal modifications in HLA-DR+ T lymphocytes (HLA-DR
`phenotype expressed by activated T cells) and in CD8+ CD57 +
`cells (known to manifest NK-like activity) in a cohort of 34
`patients with advanced ARC treated with AZT. Changes in
`levels of circulating CD4+ cells were simultaneously measured.
`The anti-retroviral activity of AZT was evaluated in terms of
`177
`
`Ex. 1053 - Page 3
`
`

`

`"Ilasgow
`ow
`London
`on
`iden
`Bethesda
`ondon
`r Harm
`row
`!bridge
`Glasgow
`don
`
`tor
`
`?ill also be
`
`:es will be
`quest this
`is at the
`
`e Editor.
`munology
`ey Green,
`71-2783).
`
`subscrirr
`it to the
`ey Mead,
`05).
`
`4SCA.
`Science
`
`nation to
`temal or
`>cientihc
`with the
`eportillg
`is paid
`chases
`
`Clin. exp. Immunol. (1990) 81, 177-182
`
`T activation marker evaluation in ARC patients treated with AZT.
`Comparison with CD4 ± lymphocyte count in non-progressors and
`progressors towards AIDS
`
`M. LEVACHER, S. TALLET, M. C. DAZZA, E. DOURNON, B. ROUVEIX & J.J. POCIDALO
`Pharmacology Laboratory, INSERM U13, Flopital Claude Bernard, Paris, France
`
`(Accepted for publication 13 February 1990)
`
`SUMMARY
`
`Reductions in the percentage and absolute number of CD4+ lymphocytes, as well as abnormally high
`levels of activated peripheral T lymphocytes (CD3 + HLA-DR+ phenotype) and an increased
`proportion of CD8÷ cells coexpressing the CD57 surface antigen (involved in natural killer activity)
`have been reported in HIV infection and associated with disease progression. We prospectively
`measured these subsets of lymphocytes in 34 patients with advanced AIDS-related complex (ARC)
`treated with azidothymidine (AZT). Peripheral blood lymphocyte phenotyping was performed
`before treatment, then at weeks 12 and 24. A striking fall in the proportion of activated T
`lymphocytes from baseline was observed (P < 0.001) at week 24. In contrast, the percentage of CD4+
`cells showed a slight and transient rise at week 12 (P < 0-05). No modification in levels of CD8+ or
`CD8+ CD57 + cells was detected during the study. Of the 34 patients, 11 developed AIDS, and 23
`remained AIDS-free during 51 weeks of follow-up. Similar patterns of change in CD4+ and HLA-
`DR+ CD3+ lymphocytes were found in the AIDS progressors and nonprogressors. Likewise, HIV
`p24 antigenaemia showed parallel decreases in both groups of patients. Although changes in CD4+
`cells, p24 antigenaemia and HLA-DR-reactive T lymphocytes were not predictive of clinical
`outcome, large differences existed between the two groups prior to the initiation of therapy. The
`short-term onset of AIDS was associated with lower CD4+ cell numbers, higher levels of serum p24
`antigen and a greater proportion of activated T lymphocytes. Our results suggest that the possible
`interest of T lymphocyte activation markers, in conjunction with conventional phenotyping, should
`be investigated further.
`
`Keywords CD4 lymphocytes activated T lymphocytes ARC AZT
`
`INTRODUCTION
`
`A reduction in the percentage and absolute number of T helper
`lymphocytes is the hallmark of HIV infection. Epidemiologic
`and clinical studies have indicated that the level of CD4+
`circulating lymphocytes is one of the most important correlates
`With disease progression (Moss et al., 1988). Serial measurement
`of CD4+ cells is thus commonly used to detect changes in
`Immune status with time, due to disease progression or the
`effects of therapeutic intervention. The clinical benefit of
`a
`zidothymidine (AZT) treatment in patients with AIDS and
`severe AIDS-related complex (ARC) has been reported in
`several trials (Fisch] et al., 1987; Dournon et al., 1988) but is
`associated only with a modest and transient increase in CD4+
`cell levels. However, this biological parameter shows large
`C
`orrespondence: Dr Maryse Levacher, Laboratoire de Pharmacolo-
`gie, INSERM U13, Hopital Claude Bernard, 10, Avenue de la Porte
`d Aubervilliers, 75944 Paris Cedex 19, France.
`
`individual variability (Taylor et al., 1989), making accurate
`assessment difficult. Several other substantial lymphocyte sub-
`set alterations have been reported in HIV-infected individuals
`(Lane & Fauci, 1985). For instance, high levels of peripheral T
`cells expressing antigenic determinants related to a state of
`activation have been described and forwarded as an indicator of
`disease stage (Prince, Arens & Kleinman, 1987; Giorgi & Detels,
`1989). Likewise, impairments of phenotypic subsets associated
`with natural killer (NK) activity suggest that these cells are
`involved in the pathogenesis of HIV infection (Hahn et al.,
`1989).
`We therefore performed a prospective study of the numeri-
`cal modifications in HLA-DR + T lymphocytes (HLA-DR
`phenotype expressed by activated T cells) and in CD8+ CD57 +
`cells (known to manifest NK-like activity) in a cohort of 34
`patients with advanced ARC treated with AZT. Changes in
`levels of circulating CD4+ cells were simultaneously measured.
`The anti-retroviral activity of AZT was evaluated in terms of
`177
`
`Ex. 1053 - Page 4
`
`

`

`r
`
`GILLES, S.,
`oning of the
`Lily. Science,
`
`1) Increased
`)cytes after
`Clin. exp.
`
`unoglobulin
`
`rs are anti-
`ritis Rheum.
`
`a
`
`0) Gamma-
`rheumatoid
`
`VANISH!, Y.,
`t88) Cloning
`' 2) receptor.
`
`r
`
`r
`
`r
`
`r
`
`r (cid:9)
`
`Clin. exp. Immunol. (1990) 81, 301-305
`
`Tumour necrosis factor-alpha and interferon-gamma production measured
`at the single cell level in normal and inflamed human intestine
`
`T. T. MACDONALD, P. HUTCHINGS,* M.-Y. CHOY, S. MURCH & A. COOKE*
`Department of Paediatric Gastroenterology, St Bartholomew's Hospital, London, and *Department of Immunology,
`Middlesex Hospital, London, England
`
`(Accepted for publication 27 February 1990)
`
`SUMMARY
`
`The spot-ELISA technique has been used to enumerate the frequency of cells secreting tumor necrosis
`factor-alpha (TNF-a) and interferon-y (IFN-y), isolated from biopsies of normal intestine and from
`biopsies of children with inflammatory bowel disease. TNF-a production was undetectable in six out
`of 12 biopsies from normal intestine and in the other six biopsies it ranged from 60 to 580 TNF-a-
`secreting cells/106 isolated intestinal cells. In contrast, cells isolated from biopsies of children with
`Crohn's disease (n = 9) all showed elevated frequencies of TNF-a-secreting cells (500-12 000 secreting
`cells/106 cells). In ulcerative colitis, four out of eight children had increased production of TNF-a and
`in children with indeterminate colitis two out of three had elevated levels. There was no correlation
`between plasma TNF-a levels and the number of intestinal cells secreting TNF-a. In controls and all
`groups of patients IFN-y-secreting cells were uncommon. These results suggest that TNF-a is an
`important mediator of inflammation in the human gut, and, furthermore, may play a role in the
`growth failure frequently seen in children with inflammatory bowel disease.
`
`Keywords tumour necrosis factor interferon-gamma Crohn's disease
`
`INTRODUCTION
`
`The aetiology of chronic inflammatory bowel disease (IBD)
`remains unknown. Whatever the initial stimulus, there is a
`general consensus that the immune system contributes towards
`the local intestinal inflammation. The normal intestinal mucosa
`is infiltrated with T cells, B cells, plasma cells, macrophages,
`dendritic cells, eosinophils and mast cells (Parrott, 1987). In
`IBD the density of the infiltrate of all of these cell types is
`increased (reviewed by Brandtzaeg, 1987). There have been
`numerous immunohistochemical studies on the changes of
`plasma cell isotype, T cell populations, macrophages and
`dendritic cells in IBD (Selby et al., 1983a, 1983b, 1984;
`MacDermott, et al., 1986; Brandtzaeg, 1987; Allison et al., 1988;
`Seldenrijk et al., 1989; Mahida et al., 1989b). There have also
`been several studies on local T cell immunoregulation and
`immunoglobulin production in IBD but these have been
`uninformative (Fiocchi, Youngman & Farmer, 1983; Elson,
`Machelski & Weiserbs, 1985; James et al., 1985). In contrast,
`there have been very few studies on local production of T cell-
`derived and macrophage-derived mediators of inflammation in
`
`IBD, although it is well established that mediators such as
`leukotrienes and prostaglandins are produced at increased levels
`(Zifroni et al., 1983; Lauritsen et al., 1988).
`The spot-ELISA technique allows quantification at the
`single cell level of the frequency of cytokine secreting cells in any
`cell population (Hutchings et al., 1989). We therefore used this
`technique to quantify the frequency of cells secreting tumour-
`necrosis factor-alpha (TNF-a) (cachectin) and interferon-
`gamma (IFN-y) in normal and inflamed human intestine. The
`choice of these two mediators was not random since children
`with IBD frequently show growth failure (Silverman, 1966;
`O'Donogue & Dawson, 1977) which may be related to the
`presence of increased levels of circulating TNF-1 as well as other
`factors, such as impaired nutrition and protein-losing entero-
`pathy. In addition, the intestinal lesion in IBD shows some of
`the phenomena associated with local production of IFN-y, such
`as increased HLA-DR expression and granuloma formation
`(Rappaport, Burgoyne & Smetana, 1951; Lockhart-Mummery
`& Morson, 1960; Selby et al., 1983a; MacDonald, Weinel &
`Spencer, 1988).
`
`MATERIALS AND METHODS
`
`Correspondence: Dr T. T. MacDonald, Department of Paediatric
`Gastroenterology, St Bartholomew's Hospital, London EC1A 7BE,
`England.
`
`Subjects
`These studies were carried out with the approval of the Hackney
`and District Health Authority Ethical Committee. Tissue was
`301
`
`Ex. 1053 - Page 5
`
`

`

`302
`
`T. T. MacDonald et al.
`
`Table 1. Clinical details of the patients with inflammatory bowel disease
`
`Patient no. (cid:9)
`
`Sex Age (years)
`
`Disease
`
`Treatment*
`
`Crohn's disease
`F
`1 (cid:9)
`
`13.3
`
`Ileum, pan-colitis
`
`2
`
`3
`4
`5
`6
`7
`8
`9
`
`Ulcerative colitis
`1
`2
`3
`4
`
`5
`
`6
`7
`8
`
`Indeterminate
`colitis
`1
`
`2
`
`3
`
`13.1
`
`11.0
`11.6
`15.0
`7.9
`9.9
`16.1
`11.6
`
`15.5
`9.4
`3.1
`11.0
`
`16.0
`
`11.2
`9.8
`17.7
`
`Ileum, severe colonic disease,
`rectal sparing
`Ileum, colon, rectal sparing
`Ileum, colon
`Ileum only
`Ileum, pan-colitis
`Ileum, right colon
`Ileum, right colon
`Ileum, right colon
`
`pancolitis
`proctitis
`pancolitis
`disease from sigmoid colon to
`caecum
`descending colon
`
`pancolitis
`pancolitis
`Left colon, rectum
`
`9.8
`
`pancolitis
`
`12.7
`
`7.1
`
`pancolitis
`normal rectum
`caecum
`
`Pred (25 mg in d.)
`Azath (25 mg in d.)
`Asacol (400 mg in d.)
`Asacol (400 mg in d.)
`
`None
`Asacol (800 mg b.i.d.)
`SZP (2000 mg t.i.d.)
`SZP (1500 mg t.i.d.)
`None
`Flexical
`Cimetidine (400 mg b.i.d.)
`
`Asacol (1200 mg b.i.d.)
`None
`SZP (600 mg t.i.d.)
`SZP (1500 mg t.i.d.)
`
`Asacol (1200 mg t.i.d.)
`Pred (2 mg b.i.d.)
`Pred (30 mg b.i.d.)
`Pred (10 mg b.i.d.)
`Asacol (800 mg b.i.d.)
`Pred (15 mg in d.)
`
`Asacol (800 mg b.i.d.)
`Pred (20 mg b.i.d.)
`SZP (750 mg t.i.d.)
`Pred (10 mg t.i.d.)
`Pred (5 mg b.i.d.)
`Asacol (800 mg b.i.d.)
`
`* Treatment during the 4 weeks prior to colonoscopy.
`Pred, prednisolone; Azath, azathioprine; SZP, salazopyrine.
`
`obtained from 20 children with IBD: nine with active Crohn's
`disease; eight with active ulcerative colitis; and three with active
`indeterminate colitis (intestinal inflammation with none of the
`characteristics of Crohn's disease or ulcerative colitis). Details
`of these patients are given in Table 1. None had prior intestinal
`surgery. As controls, 12 patients were studied in whom IBD was
`suspected (seven girls and five boys, age range 1-14.5 years).
`They were admitted for colonoscopy because of recurrent
`abdominal pain, diarrhoea, and in some cases unexplained
`weight loss. However, on clinical and histologic analysis, a
`diagnosis of IBD was excluded. None of these children has
`subsequently developed IBD and all are healthy. Clinical
`diagnosis was based on the criteria of Chong et al. (1985).
`
`Biopsies
`During investigative colonoscopy to assess the extent of the
`inflammation in patients in relapse, or in patients under initial
`investigation, two biopsies were taken from regions of the
`mucosa showing macroscopic inflammation. Biopsies were
`taken for histologic analysis from adjacent sites and confirmed
`that the tissues were inflamed. In patients with Crohn's disease,
`samples were taken from the ileum (patients 2, 5, 7) and from the
`
`ascending or transverse colon in the rest. In patients with
`ulcerative colitis, if the patient had pancolitis the biopsies were
`taken from the right colon or transverse colon, and in two
`patients with distal disease one was biopsied in the rectum and
`the other in the left colon. All biopsies from the control patients
`were made in the right colon or transverse colon. The biopsies
`were immediately placed in tissue culture medium, transported
`to the Paediatric Gastroenterology Laboratory, and the cells
`isolated by collagenase digestion as previously described (Mac-
`Donald et al., 1987). Since these cells contain numerous IgA
`plasma cells they are predominately from the lamina propria
`although some intra-epithelial lymphocytes may also be present.
`Cell yields varied from 4 x 105 to 1.5 x 106 from the two biopsies
`and there was a tendency for higher cell yields from patients with
`IBD. Plasma was also collected from some of the children at the
`time of colonoscopy and stored at —70°C.
`
`Spot-ELISA assays
`These were carried out exactly as described elsewhere (Hutch-
`ings et al., 1989). Aliquots of the cells were plated in triplicate
`or
`into microtitre wells coated with monoclonal anti-TNF-%
`IFN-y antibody. After overnight incubation the cells were
`
`Ex. 1053 - Page 6
`
`

`

`TNF-a and INF-y production in IBD
`
`303
`
`100 000
`
`10 0 00
`
`1000
`
`100
`
` mucosal cells )
`6
`
`TNF-a (SFC/ 10
`
`Fig. 1. TNF-a production by mononuclear cells isolated from intestinal
`biopsies of children with Crohn's disease (CD), ulcerative colitis (UC),
`indeterminate colitis (IC) or disease controls. Means are not given since
`only for the Crohn's disease patients did the values correspond to
`samples from a normally distributed population. Note the log scale.
`There were more TNF-a-secreting cells in Crohn's disease tissue than in
`controls (P <0.0005, Kolmogorov-Smirnov two-group test). Using
`600 SFC as a cut-off point for the upper limit of normal, there was
`significantly greater TNF-a production in ulcerative colitics (P = 0.03 ,
`x2) and patients with indeterminate colitis (P = 0.038 , x2) than controls.
`
`Table 2. Lack of association between TNF-a production in
`inflamed bowel and plasma levels
`
`Patient no. (cid:9)
`
`Crohn's disease
`1
`2
`3
`4
`5
`
`Ulcerative colitis
`1
`2
`3
`4
`
`TNF (SFC/106 cells) Plasma TNF-a
`(pg/ml)
`in the mucosa (cid:9)
`
`1280
`510
`1283
`2940
`535
`
`5400
`763
`78 660
`1730
`
`<10
`<10
`30
`18
`<10
`
`<10
`<10
`27
`150
`
`washed off and bound TNF-a or IFN-y was detected using
`rabbit anti-TNF-a or anti-IFN-y followed by alkaline phospha-
`tase-conjugated sheep anti-rabbit IgG. Antibody binding was
`visualized using 5-bromo-4-chloro-3-indolyl phosphate as
`enzyme substrate and fast blue. Blue spots per well were counted
`and the frequency of spot-forming cells/10' input cells (SFC/10')
`was then derived based on the number of cells placed into the
`wells. Results given are the mean of the triplicate cultures.
`Comparison between groups was made using the Kolmogorov-
`Smirnov two-group test or by x2 analysis with continuity
`correction factor.
`
`Plasma TNF-a levels
`These were carried out using a commercial ELISA kit with a
`lower limit of sensitivity of 10 pg TNF-a/ml of plasma (T Cell
`Sciences, Cambridge, MA).
`
`Separation of E+ and E- mucosal lymphocytes
`T cell rosettes were made using AET-treated sheep erythrocytes.
`The rosetted cells were separated by centrifugation through
`Ficoll/Paque (Pharmacia). Due to the low numbers of cells
`available from biopsies it proved impossible to recover the E+
`cells from the erythrocyte pellet by water lysis. Thus the
`comparison was made between unseparated mucosal lympho-
`cytes and the E- cells recovered from the buffy coat (between 10
`and 50% of the original cell number).
`
`RESULTS
`
`TNF-a production in normal and diseased mucosa
`In six of 12 controls, TNF-a was undetectable and in the other
`six patients, levels range was 60-580 SFC/106 mucosal cells
`(median 200 SFC, Fig. 1). In contrast, patients with Crohn's
`disease had between 510 and 12 000 SFC/106 cells (median 1280
`SFC). Four of the eight patients with ulcerative colitis had no
`TNF-a-secreting cells, and in the other four, levels ranged from
`620 to 78 000 SFC/106 mucosal cells. In three patients with
`indeterminate colitis, TNF-a-secreting cells were 480, 1700 and
`1730 SFC/10" intestinal cells.
`No correlation was apparent between local production in
`the gut of TNF-a and plasma levels (Table 2). Of nine patients
`tested with raised levels in the gut, five had no detectable plasma
`TNF-a and in the others the values ranged from 18 to 150 pg/ml.
`Comparing the frequency of TNF-a-secreting cells in unsepar-
`ated cells and the E- cells, it was found that in four patients,
`most of the activity appeared to be associated with the E+ cells
`(Table 3). In another patient, the E- cells had around 50% of the
`activity of the unseparated cells. It is unlikely that E rosetting
`non-specifically removes TNF-a-secreting cells, since in a single
`patient with ulcerative colitis the E- cells were enriched for
`TNF-a-secreting cells compared with unseparated cells.
`
`IFN-y production in normal and inflamed mucosa
`In contrast to the results with TNF-a, IFN-y-secreting cells were
`uncommon in all of the patient groups studied. Four out of nine
`Crohn's patients showed some activity (Table 4); however, two
`out of 11 controls tested also had low levels of IFN-y-secreting
`cells.
`
`ith
`re
`NO
`nd
`its
`ies
`ed
`lls
`rc-
`
`ria
`at.
`ies
`ith
`he
`
`:h-
`Ste
`or
`;re
`
`Ex. 1053 - Page 7
`
`(cid:9)
`

`

`4
`
`304
`
`T. T. MacDonald et al.
`
`Table 3. TNF-a secreting cells in unseparated and
`E — mucosal cells
`
`TNF-a (SFC/106 cells)
`
`Unseparated
`
`E —
`
`Crohn's disease
`Crohn's disease
`Indeterminate colitis
`Indeterminate colitis
`Ulcerative colitis
`Ulcerative colitis
`
`5920
`12 066
`1730
`480
`78 660
`1710
`
`60
`5700
`60
`0
`2820
`10 510
`
`Table 4. IFN-y production is not a prominent feature of
`inflammatory bowel disease
`
`Patients (cid:9)
`Controls (cid:9)
`Crohn's disease
`Ulcerative colitis
`Indeterminate colitis
`
`Positives
`No. (cid:9)
`positive/total (SFC/106 cells)
`
`2/11
`4/9
`2/7
`2/3
`
`93, 113
`22, 53, 370, 1500
`20, 60
`90, 430
`
`DISCUSSION
`
`To our knowledge this is the first time in which cells secreting
`inflammatory mediators have been quantified in normal and
`diseased human intestine. The major observation is that the
`frequency of cells secreting TNF-a is increased in the mucosa in
`IBD. All of the Crohn's patients, half of the ulcerative colitis
`patients and two out of three patients with indeterminate colitis
`showed raised levels. We cannot yet explain the heterogeneity in
`the ulcerative colitis patients, since all the biopsies were taken
`from apparently inflamed mucosa, and the results did not
`correlate with steroid treatment.
`There has been interest in TNF-a as a mediator of inflamma-
`tion in recent years (reviewed by Beutler & Cerami, 1987;
`Cerami & Beutler, 1988), and it clearly has many diverse effects.
`However, we would like to comment briefly on features of its
`biological activity of relevance to gastrointestinal disease and its
`systemic sequelae.
`TNF-a infused systemically into rats causes intestinal
`haemorrhage and necrosis (Sun & Hsueh, 1988). This is due to
`break-down of the endothelium in the intestinal vascular
`network caused by TNF-induced pro-coagulant activity of the
`endothelial cells and increased neutrophil adherence and media-
`tor release (Cerami & Beutler, 1988). TNF-a is also angiogenic
`and may help in the healing of diseased mucosa (Frater-
`Schroder et al., 1987).
`TNF-a has also been shown to be important in the
`generation of the granulomatous response to Mycobacteria in
`mice (Kindler et al., 1989). TNF-a acts in an autocrine fashion,
`being secreted by activated macrophages, and inducing other
`macrophages to differentiate into epithelioid cells. A macro-
`phage influx is a feature of Crohn's disease and to a lesser extent
`
`ulcerative colitis (Selby et al., 19831); Allison et al., 1988.
`Seldenrijk et al., 1989; Mahida et al., 1989b); however, it is still
`unknown why granulomas are seen only in Crohn's disease.
`TNF-a also has cachectic activity (Oliff et al., 1987; Tracey
`Manogue et al., 1988; and may contribute to weight loss
`frequently seen in children with IBD (as do other factors, such as
`poor nutrition).
`Raised plasma C-reactive protein (CRP) levels are also
`associated with IBD (Walker-Smith, 1988). Infusion of graded
`doses of TNF-a into patients leads to a dose-dependent increase
`in plasma CRP levels (Michie et al., 1988).
`In preliminary experiments we attempted to identify the
`mucosal cell type secreting the TNF-a. Our results were
`equivocal in that in some patients all of the activity appeared to
`be in the T cell fraction of mucosal cells, in another patient
`activity was associated with the non-T cell fraction, and in
`another patient, both fractions had activity. There is however no
`a priori reason to assume that only macrophages secrete TNIF-2
`in inflamed bowel since it is now well established that T cells can
`also secrete this mediator (Cuturi et al., 1987; Pawlec et al.,
`1989). It is likely that there are numerous mediators being
`produced in inflamed bowel. For example, it has recently been
`shown that interleukin-1 secretion by lamina propria macro-
`phages is increased in patients with IBD (Mahida, Wu &Jewell,
`1989a). It is also unlikely that TNF-a production is specific for
`IBD and that as more enteropathies are studied TNF-1
`production may be found to be a common feature.
`Class II MHC expression on epithelial cells is increased in
`IBD (Selby et al., 1983a), a feature usually attributed to local
`production of IFN-y by activated T cells (MacDonald et al.,
`1988). However, a striking feature of our study was the relative
`infrequency of cells secreting IFN-y in diseased mucosa. This
`was surprising in view of the large numbers of activated T cells in
`the mucosa in Crohn's disease (Mahida etal., 1988; Choy etal.,
`1990). However, a single previous report also showed that
`mitogen induced IFN-y secretion was lower in mucosal lympho-
`cytes from patients with Crohn's disease compared with con-
`trols (Ouyang et al., 1988).
`
`ACKNOWLEDGMENTS
`A. Cooke and T. T. MacDonald are Wellcome Senior Lecturers.
`M. Choy is supported by a grant from the National Foundation for
`Ileitis and Colitis. This work is also supported by the MRC, and Crohn's
`Disease in Childhood Research Association.
`
`A.P. &
`
`REFERENCES
`ALLISON, M.C., CORNWALL, S., POULTER, L.W., DHILLON,
`POUNDER, R.E. (1988) Macrophage heterogeneity in normal colonic
`mucosa and in inflammatory bowel disease. Gut, 29, 1531.
`BEUTLER, B. & CERAMI, A. (1987) Cachectin: more than a tumor necrosis
`factor. N. Engl. J. Med. 316, 379.
`BRANDTZAEG, P. (1987) The B cell system. In Food Allergy and
`Intolerance (ed. by J. Brostoff & S. B. Challacombe) p. 118. Bailliere
`Tindall, London.
`in
`CERAmi, A. & BEUTLER, B. (1988) The role of cachectin/TNF
`endotoxic shock and cachexia.