Immunological study of the rectal mucosa of men with and without human immunodeficiency virus infection - Gut

 
CONTINUE READING
Gut: first published as 10.1136/gut.28.12.1619 on 1 December 1987. Downloaded from http://gut.bmj.com/ on November 13, 2021 by guest. Protected by copyright.
                                                                                                 Gut, 1987, 28, 1619-1624

Immunological study of the rectal mucosa of men with
and without human immunodeficiency virus infection
P E BISHOP, A McMILLAN, AND H M GILMOUR
From the Department of Genito-Urinary Medicine and Department of Pathology, University of Edinburgh,
Edinburgh

SUMMARY     Biopsies of rectal mucosa were taken from 81 men and stained using cytochemical
methods for B and T lymphocytes, T cell subsets, immunoglobulin containing plasma cells and
mucosal mast cells. The patients studied included human immunodeficiency virus (HIV) infected
and non-infected heterosexual and homosexual men, and homosexual men with rectal gonorrhoea.
There were increased numbers of T lymphocytes in the lamina propria of the rectum in HIV
infected individuals regardless of whether the infection had been acquired through anal intercourse
or intravenous drug use. This increase resulted from a marked increase in the numbers of CD8+
suppressor T cells, there also being a reduction in the numbers of CD4+ helper T cells. In non-HIV
infected men with rectal gonorrhoea there were increased numbers of CD8+ T cells but no
significant difference in numbers of CD4+ cells. No difference was seen in numbers of
immunoglobulin containing plasma cells or mucosal mast cells between HIV infected and non-
infected men.

In Western Europe and the United States of America               infected intravenous drug users (all with persistent
sexually active homosexual men constitute the pop-               generalised lymphadenopathy, without history of
ulation group at most risk for acquiring the human               homosexual contact) and 67 consecutive homosexual
immunodeficiency virus (HIV).'12 Most studies have               male patients attending the department (Table 1).
shown a clear association between anal intercourse,              All but three of the healthy homosexual men had
particularly penoreceptive, and acquisition of the               been the recipient partners during anal intercourse.
virus.34 As the rectum seems to be the portal of entry             After taking a careful history and completing a
of HIV it is surprising that there have been few                 general physical examination, material for micro-
reports of the immunohistology of the rectum in                  biological examination was obtained.' A sigmoid-
homosexual men. The present study was undertaken                 oscope, lightly lubricated with K-Y jelly (Johnson
to discover if the cellular content of the lamina                and Johnson, Slough, UK) was passed and using
propria of the rectums of sexually active homosexual             Patterson's forceps at least two biopsies were taken.
men was different from that of heterosexual men,
and to assess the effect of gonococcal infection on the          LABORATORY METHODS
lymphocyte composition of the rectal mucosa.                              One biopsy was carefully orientated on a glass slide,
                                                                          fixed in buffered formol sublimate and processed for
Methods                                                                   paraffin sections for histological evaluation. A
                                                                          second similarly orientated biopsy was snap frozen in
PATIENTS AND CLINICAL METHODS                                             liquid nitrogen and subsequently embedded in OCT
The study was approved by the Ethics Committee of compound (Miles Scientific, Naperville, USA) for
the Lothian Health Board. Eighty one men were the preparation of cryostat sections. The block was
studied: nine healthy heterosexual men, five HIV sectioned immediately or stored at -70°C for up to
                                                                          one week. A third biopsy, obtained from 24 con-
Address for correspondence: Dr A McMillan, Department of Genito-Urinary   secutive patients, was fixed in Carnoy's fluid and
Medicine, The Royal Infirmary, Edinburgh EH3 9YW.                         processed for paraffin sections for staining for mast
Received for publication 8 May 1987.                                      cells.
                                                                       1619
Gut: first published as 10.1136/gut.28.12.1619 on 1 December 1987. Downloaded from http://gut.bmj.com/ on November 13, 2021 by guest. Protected by copyright.
1620                                                                                              Bishop, McMillan, and Gilmouir
Table 1 Diagnoses made in the 67 hom0osexual men                       ELISA or indirect immunofluorescence methods
studied                                                                with confirmation of positive results by western
                                                                       blotting.9
lIiagniosis                                     Metz   affected (it)
                                                                       STATISTICAL ANALYSIS
Syphilis (early)                                 2                     The data obtained on the lymphocyte subset compo-
Rectal gonorrhoca                                9                     sition of the lamina propria were normally distri-
Non-gonococcal urethritis                        I
Chiamydial infection of rectum                   1                     buted and Student's t test was used in their analysis.
Anorectal herpes simplex infection               2                     As the numbers of mucosal mast cells and immuno-
Perianal condylomaita acuminata                 17                     globulin containing plasma cells were not normally
Amochiasis
Giardiasis                                                             distributed, Wilcoxon's rank-sum test was used. The
Enterobiasis                                                           x2 test was used to compare the histological grading of
HIV infection                                    9t                    the rectal mucosa.
No infection identified                         36
                                                                       Results
* Some men had concurrent infections; t Eight men h.ad persistent
generalised lymphadenopathy. the other had no clinical evidence of
HIV infection.                                                         HISTOL OGICAL GRADING (Table 2)
                                                                       The histology of the rectum was normal (grade A) in
IMMUNOCYTOCHEMICAL METHODS                                             the 14 heterosexual men studied, including the five
Cryostat sections at 4 [tm were cut at two levels and                  intravenous drug users seropositive for HIV. A mild,
stained for lymphocyte subsets using an indirect                       non-specific proctitis was found in five of 14 HIV
immunoperoxidase method6 with the following                            infected and six of 36 non-infected homosexual men.
monoclonal antisera: Pan-B/M708 (CD22, mature B                        There was no significant difference in the prevalence
cells, Dako Ltd., High Wycombe, UK), anti-Leu 4                        of inflammatory changes between HIV infected and
(CD3a, mature T cells), anti-Leu 2a (CD8, sup-                         non-infected men (X'= 1 79; p>0. 10).
pressor/cytotoxic T cells) and anti-Leu 3a (CD4,
helper/inducer T cells) (all from Becton Dickinson,                    I YMPHOCYTE SUBPOPUI ATIONS WITHIN THE
Laboratory Impex, Twickenham, UK).                                     L AMINA PROPRIA (Figure)
   Paraffin sections at 3-4 sim taken at two levels
from 15 consecutive blocks were stained for plasma                     CD22' B cells
cells containing IgA, IgG and IgM using a peroxidase-                  Using the antiserum M708, which is broadly reactive
antiperoxidase (PAP) method.7                                          with mature B cells, no staining was seen in the
                                                                       lamina propria. All B cells reactive with this anti-
STAINING FOR MAST CELLIS                                               serum were confined to organised lymphoid aggre-
Mucosal mast cells were stained with Astra blue                        gates.
(BDH Ltd., Poole, UK) and safranin as described by
Strobel et al.'                                                        Table 2 Histological grading of rectal biopsies in relation to
                                                                       HIV infection and gonorrhoea
HISTOMETRY
Cells were counted using an eyepiece graticule                                                               Histological grade
marked with a lOx 10 grid. The sections were
examined under high power magnification (objective                     Group                                 A           B           Total
 x40, eyepiece x 10) and one edge of the grid aligned
with the basement membrane of the surface epi-                         Non-infected. non-intravcnous
                                                                       drug-using heterosexual men             9         t)           9
thelium. At this magnification the grid edge covered
                                                                       IIIV-infected heterosexual
a length of 0-283 mm, area 0 08 mm2. Where                             intravenous drug uscrs                  5         (            5
orientation of the sections allowed, five high power                   Non-infceted homoscxual men           311         5           36
fields were counted (minimum three) by two                             IiIV infected homoscxual men            9'        5           14
observers independently. Areas including, or im-                       hiomosexual men with rectal
mediately adjacent to lymphoid aggregates were                         gonorrhoea but no scrological
avoided. Interobserver variation was
Gut: first published as 10.1136/gut.28.12.1619 on 1 December 1987. Downloaded from http://gut.bmj.com/ on November 13, 2021 by guest. Protected by copyright.
Rectal lymphocytes and HIV infection                                                                                      1621

          CD3a+ cel Is                                                             CD4+ cel Is
 80-

 60-

                                                        4
 40-
                                    I0

 20-                      4
                                                            1-
  0              A         B            C               D
 80-                                                                    8-
          CD8+ cells                                                               CD4+: CD8 cell ratio

 60-                                                                    6-

 20-

  0             WZZ

                 A        B
                                   Iz

                                   rz

                                        C
                                            fz         wZgL-
                                                        D
                                                                 Al
                                                                        4-

                                                                        2-

                                                                        0   A
                                                                                        I
                                                                                        B
                                                                                         4;
                                                                                       rl~Z      v le

                                                                                                  C
                                                                                                        -sz A

                                                                                                             D
                                                                                                                o     o

Figure Numbers of T cells per unit area (0-08 mm2 ) and ratio of CD4 :CD8+ cells in the lamina propria of the rectum.
A=non-infected heterosexual men (n=5); B=non-infected homosexuial men (n=12); C=HIV infected homosexual men
(n =9); D=HIV infected heterosexual men (n=5); *errorbar

CD3a T cells                                                          HIV infected homosexual men (16-1 ± 6-3) was
Within the unit area of 0-08 mm) there was no                         lower than that of the non-infected homosexual men
significant difference in the number of T cells be-                   (30.3 ± 9-4): this difference was highly significant
tween non-infected heterosexual (44-0 ± 11.8) and                     (t=3-91; p0- 1). Similarly, the mean numbers of T cells in the                non-infected (47.6 ± 28-1) heterosexual men
mucosa of HIV infected homosexual men (67.3 ±                         (t=2.75; p0-5). The mean                 CD8' cells
numbers of CD3a' cells in the lamina propria of HIV                   The mean number of CD8+ cells in the lamina
infected homosexual men and intravenous drug users                    propria was significantly greater in the HIV infected
were significantly higher than those of the non-                      (50-4 ± 26-1) than in the non-infected (8-2 ± 2-8)
infected men (t=5.15; p
Gut: first published as 10.1136/gut.28.12.1619 on 1 December 1987. Downloaded from http://gut.bmj.com/ on November 13, 2021 by guest. Protected by copyright.
1622                                                                                       Bishop, McMillan, and Gilmtnoiur
CD4': CD8 cell ratio                                            Discussion
As can be seen from the Figure, the CD4+: CD8'
cell ratio was significantly lower in the two HIV               This study has shown that in HIV infected indi-
infected groups (0.4 ± 0(3) compared with the non-              viduals, regardless of how the infection had been
infected groups (3-9 ± 1-1) (t=11 3; p). I                       p>0. I
Gut: first published as 10.1136/gut.28.12.1619 on 1 December 1987. Downloaded from http://gut.bmj.com/ on November 13, 2021 by guest. Protected by copyright.
Rectal lymphocytes and HlV infection                                                                              1623

 cells between Rodgers and colleagues"' and our study        potentiation of eosinophil and macrophage cyto-
 is not immediately apparent. Among sexually active          toxicity." Although increased numbers of mucosal
 homosexual men, infection with Epstein-Barr virus           mast cells were noted in men with AIDS associated
 (EBV) and cytomegalovirus (CMV) is universal,'2             diarrhoea," we did not find a difference in the median
 and there are data to suggest that HIV infection may        numbers of cells per unit area between HIV and non-
 reactivate EBV.'" Reactivation of these viruses is          HIV infected men.
 sometimes associated with increased numbers of                 Secretory IgA is important in the host defence
 CD8-bearing lymphocytes in the peripheral blood.'           against infection at mucosal surfaces."1 As CD4+ cells
 Unfortunately facilities for viral culture from our         play a part in the regulation of IgA responses,'7 HIV
 patients were not available. Although the number of         infection might result in impaired mucosal immunity.
 CD8' cells in the peripheral blood of homosexual            Indeed, intestinal infection of such individuals
 men infected with HIV is often greater than normal,'        with organisms such as Crypto-sporidium spp that
 it is difficult to know whether this represents a direct    generally produce a transient diarrhoeal illness is well
 effect of that infection or the result of reactivation of   recognised.2" Although we did not find a significant
 latent viruses such as CMV or EBV. After acute HIV          difference in the median numbers of IgA containing
 infection, there is an increase in the number of CD8        plasma cells within the lamina propria of HIV
 cells in the peripheral blood.4 The only patient in our     infected and non-infected men, a functional impair-
 series who had a mononucleosis like illness followed        ment of IgA secretion is not precluded.
 by seroconversion had the most marked infiltration             In conclusion, our findings confirm that abnor-
 of the lamina propria with CD8 cells of all the             malities of lymphocyte subset distribution within the
 patients studied. The mean number of CD4' cells (33         lamina propria of the rectums of HIV infected men
 per unit area) was the highest of the HIV infected          are common, but their significance with respect to
 homosexual men but within one standard deviation            mucosal immunity is still uncertain.
 of the mean value for non-infected men. As a result
 of the large increase in CD8X cells the CD4':CD8,           We would like to thank Dr A E Dewar, Department
 ratio was low (0-4). None of the other patients,            of Pathology, University of Edinburgh, for her
 however, were studed at this stage of infection             helpful advice. This work was supported by Scottish
 (median duration of lymphadenopathy 12 months,              Home and Health Department, Biomedical Re-
 range five to 24 months).                                   search Committee grant K/MRS/50/C685.
    Although in most studies receptive anal inter-
 course has been shown to be a risk factor for the           References
 development of HIV antibody,4 this view was chal-            1 Cheingsong-Popov R, Weiss RA, Dalgleish A, et al.
 lenged by Weber et alP who did not find an associa-            Prevalence of antibody to human T-lymphotropic virus
 tion between anal intercourse and seropositivity for           type III in AIDS and AIDS-risk patients in Britain.
 HIV. These data, however, suggested that concur-               Lanicet 1984: ii: 477-80.
 rent or recent sexually transmitted disease might be         2 Gallo RC, Salahuddin SZ, Popvic M, et al. Frequent
 an important cofactor in developing HIV antibody.              detection aind isolcation of cytopathic retroviruses
 They postulated that acute infection may lead to T             (HTLV-III) from patients with AIDS and att risk for
 cell activation with viral replication leading to a            AIDS. Science 1984; 224: 500-3.
 humoral response. In about 40% of patients gono-             3 Melbye M, Biggar RJ, Ebbesen P, et atl. Seroepidemi-
 coccal infection of the rectum is associated with              ology of HTLV-III antibody in Danish homosexuall
 increased numbers of chronic inflammatory cells in             men: prevalence, transmission and disease outcome. Br
                                                                Med J 1984; 289: 573-5.
 the lamina propria. We have shown that this results          4 Goedert JJ, Sarngadharan MG, Biggar RJ, et cial.
 from infiltration with CD8' lymphocytes. Class II              Determinants of retrovirus (HTLV-III) antibody aind
 MHC (HLA-D region) antigens were expressed on                  immunodeficiency conditions in homosexual men.
 some of these cells (data not shown) but, as a clear           Lanicet 1984; ii: 711-6.
 differentiation between stained lymphocytes and              5 McMillan A, Lee FD. Sigmoidoscopic and microscopic
 macrophages was impossible, conclusions regarding              appearance of the rectal mucosa in homosexual men.
 the activation of the lymphocytes cannot be drawn.             Glit 1981;:22: 1035-41.
 Although many cells in the lamina propria expressed          6 Salter DM,  Krajewski AS, Dewar AE. Immuno-histo-
 class II MHC antigens, these antigens were only                chemical staining of non-Hodgkin's lymphomia with
                                                                monoclonal antibodies specific for the leucocyte com-
 expressed by the surface epithelium when overlying             mon antigen. J Paithol 1985; 146: 345-53.
 organised aggregates of lymphoid tissue."                    7 McMillan A, McNeillage G, Gilmour HM, Lee FD.
    Mast cells are involved in immediate and delayed            Histology of rectal gonorrhoea in men, with a note on
 onset hypersensitivity reactions, regulation of                anorectal infection with Neisseria nineiilgitidis. J (lin
 immune responses, direct cytotoxicity and the                  Pathol 1983; 36: 511-4.
Gut: first published as 10.1136/gut.28.12.1619 on 1 December 1987. Downloaded from http://gut.bmj.com/ on November 13, 2021 by guest. Protected by copyright.
1624                                                                                     Bishop, McMillan, atid Gilmour

 8 Strobel S, Miller HRP, Ferguson A. Human intestinal              retrovirus infection. Definition of a clinical illness
   mucosal mast cells: evaluaition of fixation and staining         associated with seroconversion. Lancet 1985; i: 537-40.
   techniques. J Clin Pathol 1981; 34: 851-8.                 15    Weher JN, McCreainer A, Berrie E, et al. Factors
 9 Peutherer JF, Edmond E, Simmonds P, Dickson JD,                  aiffecting seropositivity to human T-cell lymphotropic
   Bath GE. HTLV-1II antibody in Edinburgh drug                     virus type III (HTLV-III) or lymphadenopathy asso-
   addicts. Lancet 1985; ii: 1129-30.                               ciiated virus (LAV) aind progression of disease in sexual
10 Rodgers VD, Fassett R, Kagnoff MF. Abnormalities in              partners of patients with AIDS. Genitourin Med 1986;
   intestinal mucosal T-cells in homosexual populations             62: 177-80.
   including those with the lymphadenopathy syndrome          16    Bishop PE, McMillan A, Gilmour HM. A histological
   and acquired immunodeficiency syndrome. Gastro-                  and immunocytochemical study of lymphoid tissue in
   enterology 1986; 90: 552-8.                                      rcctal biopsies from homosexual men. Hivsopathology.
11 Janossy G, Pinching AJ, Bofill M, et al. An immuno-              (In press)
   histological approach to persistent lymphadenopathy        17    Elson CO, Kagnoff MF. Fiocchi C. Betus AD, Targan
   and its relevance to AIDS. Clin Exp Immntnol 1985; 59:           S. Intestinal immunity ind inflammattion: recent pro-
   257-66.                                                          gress. Gastroenterologv 1986; 91: 746-68.
12 Detels R, Visscher BR, Fahey JL, et al. The relation of    18    Kotler DP, Gaetz HP. Lange M, Klein EB, Holt PR.
   cytomegalovirus and Epstein-Barr virus antibodies to             Enteropathy associated with the acquired immuno-
   T-cell subsets in homosexually active men. JAMA 1984;            deficiency syndrome. Annii Internt Med 1984; 101:
   251: 1719-22.                                                    421-8.
13 Rinaldo CR, Kingsley LA, Lyter DW, et al. Association      19    Waldman RH, Gainguly R. Immunity to infections on
   of HTLV-111 with Epstein-Barr virus infection and                secretory surfaces. J lnJect Dis 1974; 130: 419-40.
   abnormalities of T-lymphocytes in homosexual men. J        20)   Soave R, Danner RL, Honig CL, ei al. Crypto-
   Infect Dis 1986; 154: 556-61.                                    sporidiosis in homosexuial men. Annii Intternt Med 1984;
14 Cooper DA, Gold J, Maclean P, et al. Acute AIDS                  100: 504-11.
You can also read