Phagocytosis of mycobacterium tuberculosis or particulate stimuli by human monocytic cells induces equivalent monocyte chemotactic protein-1 gene expression

PHAGOCYTOSIS OF MYCOBACTERIUM TUBERCULOSIS OR PARTICULATE STIMULI BY HUMAN MONOCYTIC CELLS INDUCES EQUIVALENT MONOCYTE CHEMOTACTIC PROTEIN-l GENE EXPRESSION Jon S. Friedland, Robin J. Shattock, George E. Griffin Phagocytosis of Mycobuctetium tuberculosis by human monocytes or macrophages is classically followed by granuloma formation in vivo. Granuloma are comprised of cells of the monocyte lineage together, in many instances, with antigen-specific T lymphocytes. Development of granuloma depends upon recruitment of both cell types, but recruitment of monocytes is pivotal as these cells secrete anti-mycobacterial cytokines and IL-S, a T cell chemoattractant. We have therefore investigated gene regulation of Monocyte Chemotactic Protein 1 (MCP-l), an important monocyte chemotactic cytokine, following phagocytosis of particulate material (latex beads and zymosan) and live M. tuberculosis by two human monocytic cell lines. In THP-1 cells and phenotypically more differentiated Mono Mac 6 cells, MCP-1 mRNA accumulation was first detectable by Northern analysis at 4 hours and increased over 24 hours. Magnitude and kinetics of MCP-1 gene expression was independent of the biochemical nature of the phagocytic stimulus, M. tuberculosis strain virulence or pre-treatment with anti-TNF. In contrast to the uniform effect of different phagocytic stimuli on MCP-1 gene expression, we have shown that M. tuberculosis but not latex or zymosan, increased IL-8 gene expression, a chemotactic agent for T cells. In additional experiments with THP-1 cells infected with human immunodeficiency virus (HIV), viral infection did not alter MCP-1 gene expression following phagocytosis. MCP-1 gene expression appears to be a conserved antigen-independent response of human monocytic cells which is activated following particulate phagocytosis. MCP-1 gene expression may thus be involved in recruitment of monocytes during granuloma formation. However, MCP-1 gene expression is not altered by HIV infection and cannot account for the incomplete granuloma formation found in AIDS patients co-infected wifh tuberculosis.

Phagocytosis

of intracellular pathogens such as by monocytes or fixed tissue macrophages in vivo is the initial stimulus that classically precedes granuloma formation. Development of such granuloma is known to be dependent upon the action of cytokines. For example, TNF has Mycobacterium

tuberculosis

From the Division of Communicable Diseases, St. George’s Hospital Medical School, London. SW17 ORE, UK. Address correspondence to: Dr Jon Friedland, Division of Communicable Diseases, St. George’s Hospital Medical School, Cranmer Terrace, London, SW17 ORE, UK. Received 14 August 1992; accepted for publication 16 November 1992 0 1993 Academic Press Limited 1043-4666/93/020150+07 $08.00/O KEY WORDS: Monocyte monocytelMycobacterium

150

Chemotactic tuberculosis

Protein-l/phagocytosis/

been shown to be central to the process of granuloma formation in vivo in animal models.lJ However, the initial prerequisite for granuloma formation is recruitment of monocytes, present in all granuloma, and T lymphocytes which recognize specific antigens from a phagocytosed micro-organism when these are presented on the monocyte cell surface in association with the MHC. Recruitment of monocytes to the site of granuloma formation in tuberculosis is pivotal because these cells secrete both IL-S, a chemotactic agent for T lymphocytes3 and cytokines such as IL-6 which have direct anti-mycobacterial action.4 Thus the pattern and sequence of cytokine production and release will be important in the co-ordination of granuloma formation. Monocyte Chemotactic Protein 1 (MCP-1) has been characterized from human monocytic THP-1 cells5 and was shown to be a potent chemotactic agent CYTOKINE,

Vol.

5, No. 2 (March),

1993: pp 150-156

Phagocytosis induces MCP-1 gene expression / 151

for monocytes.6 Monocytes and macrophages express MCP-1 gene following adhesion,7 and in response to bacterial lipopolysaccharide (LPS) and cytokines such as TNF and IL-l.8 MCP-1 gene activation has been described in many non-immune cells including endothelial cells,9 fibroblastsla and pulmonary type II epithelial cells.il Such non-immune cells are stimulated by products of macrophage activation such as TNF and IL-l. Monocyte recruitment to specific sites is thus mediated by cytokine release from both tissue macrophages and indirectly via the effects of macrophage-derived cytokines on bystander cells. We were interested to determine whether phagocytosis of M. tuberculosis by human monocytic cells was a specific stimulus for MCP-1 gene expression and hence possibly involved in the development of granuloma in the host response to M. tuberculosis. The immune response to infection caused by M. tuberculosis is currently not well understood although phagocytosis of this pathogen by human monocytes does result in TNF release.12 Detailed knowledge of this phenomenon is important for the understanding of clinical tuberculosis from which there were 3 million deaths in 1990.13 Tuberculosis is increasing in frequency on a worldwide basis and this increase in both developing and developed countries is associated with human immunodeficiency virus (HIV) infection. In patients suffering from the acquired immunodeficiency syndrome (AIDS) granuloma formation may be incompletei4 and we were therefore interested to determine whether there was reduced MCP-1 gene expression in HIV infected human monocytes. In addition, we investigated the hypothesis that reduced activation of MCP-1 gene expression might be a determinant of the virulence of M. tuberculosis since delay in monocyte recruitment might potentially contribute to severity of infection. Finally, we examined the role of TNF secretion following phagocytosis on MCP-1 gene expression since TNF has been shown to increase MCP-1 gene expression and is important in vivo in granuloma formation in mycobacterial infection.*

at 4 hours, progressively increasing over a 24 hour period. The kinetics and magnitude of MCP-1 mRNA accumulation in these cells were very similar following phagocytosis of zymosan granules or virulent M. tuberculosis strain H37-Rv (Fig. 1). In contrast, IL-8 gene expression was increased 3-fold in THP-1 cells 24 hours after phagocytosis of M. tuberculosis but not raised above constitutive levels following phagocytosis of latex or zymosan (Fig. 1). Prolonged experiments over 5 days revealed that following phagocytosis of M. tuberculosis by THP-1 cells MCP-1 gene expression remained persistently increased (Fig. 2). Similar data were obtained in experiments involving the other particulate stimuli, zymosan and latex (data not shown). MCP-1 mRNA Accumulation Following Phagocytosis by Mono Mac 6 Cells The kinetics of MCP-1 gene expression following phagocytosis of latex (data not shown), zymosan and M. tuberculosis by the phenotypically more mature Mono Mac 6 cells were very similar to that observed in THP-1 cells. MCP-1 mRNA accumulation was first observed 4 hours after phagocytosis and continued to increase over 24 hours (Fig. 3). In contrast, phagocytosis of M. tuberculosis but not the other particulate stimuli resulted in a 2-3 fold increase in IL-8 mRNA with maximal levels after 224 hours which returned to baseline within 24 hours. M. tuberculosis Strain Virulence Expression in THP-1 Cells

MCP-1 mRNA Accumulation Following Phagocytosis by THP-1 Cells The MCP-1 gene was not constitutively expressed in THP-1 cells examined by Northern analysis and was not upregulated in control experiments using tissue culture medium alone which were performed to demonstrate that LPS contamination was not influencing results. Phagocytosis of inert latex beads results in accumulation of MCP-1 mRNA apparent

Gene

We next investigated the consequence of phagocytosis of either virulent (H37-Rv) or avirulent strains (H37-Ra) of M. tuberculosis by THP-1 cells. Although these two strains of M. tuberculosis have markedly different virulence in rodents,15 there were no detectable differences in the kinetics or magnitude of MCP-1 mRNA generation following phagocytosis of either strain by THP-1 cells (Fig. 4). Effect of Pre-treaIment mRNA Accumulation

RESULTS

andMCP

with Anti-TNF

on MCP

Pre-treatment of THP-1 cells with 0.1 nM monoclonal antibody directed against TNFa did not alter kinetics or magnitude of MCP-1 mRNA accumulation following phagocytosis of the three different particulate stimuli investigated. Increased MCP-1 gene expression followed phagocytosis of M. tuberculosis H37-Rv in the presence of anti-TNF as illustrated in Fig. 4. Similarly, pre-treatment with monoclonal antibody did not affect TNF and IL-8 gene expression following phagocytosis of latex or zymosan or M. tuberculosis by these cells (data not shown).

152 I Friedland

MCP mRNA (% Maw. Scanner Reading)

CYTOKINE,

et al.

Vol.

5, No. 2 (March

1993:

1X-156)

loo 90 8. 70 60 50

Figure 1. MCP-1 gene expression in THP-1 cells is similarly upregulated at 4 hours and increases over 24 hours following phagocytosis of inert latex beads or zymosan or M.tuberculosis strain H37-Rv.

30 20 10 n l-l Time

(h).

0

1

2

4

8

24

0

1

2

4

8

24

0

1

2

4

MCP:

Zymosan

However, anti-TNF did neutralize supernatant TNF bioactivity, assayed using the WEHI 164 cell line.16

Effect of Pre-treatment with Cycloheximide and Actinomycin D on MCP-1 GeneExpression in THP-I cells Following Phagocytosis Pre-treatment of THP-1 cells by cycloheximide, a protein synthesis inhibitor, induced MCP-1 gene expression even in unstimulated cells (data not shown). MCP-1 gene expression 16 hours after phagocytosis of particulate stimuli by cells treated with cycloheximide, was similar or increased when compared to gene expression following phagocytosis in untreated monocytes (Fig. 5). In contrast, pre-treatment with actinomycin D which prevents transcription of messenger RNA, completely blocked MCP-1 mRNA accumulation (Fig. 5).

MCP-I GeneExpression Following Phagocytosis by THP-1 Cells Pre-infected with HIV Infection of THP-1 cells with HIV strain RF resulted in reverse transcriptase activity and detectable p24 antigen in culture medium detectable 10 days after infection. HIV infection of THP-1 cells did not alter either magnitude or kinetics of MCP-1 gene expression. As in non-HIV infected cells, MCP-1 mRNA was first detected at 4 hours and increased over the 24 hour period following phagocytosis by THP-1 cells of any one of the three particulate stimuli studied (data not shown).

M. tuberculosis H37-RV

8 24

Both autoradiographs of the MCP-1 mRNA and their graphed representation following computerized densitometric analysis are shown. In contrast, IL-8 gene expression at 24 hours is increased following phagocytosis of M. tuberculosis H37-Rv but not other particulate stimuli (autoradiograph shown). Probing with p-actin and the equivalent 28.7 and 18s ribosomal RNA bands indicate equal loading of total RNA on gels. All figures show results from a single experiment and are representative of at least three further independent experiments.

DISCUSSION The series of experiments described in this paper demonstrate that the process of phagocytosis upregulated MCP-1 gene expression in the two human

MCP:

@actin:

28l18 s rRNA:

TIME

(days)

Figure 2. MCP-1 gene expression is persistently following phagocytosis of M. tuberculosis strain cells. Similar data were particulate stimuli.

raised for five days H37-Rv by THP-1

obtained following phagocytosis p-actin and rRNA are equivalent.

of the

other

Phagocytosis

rnonocyte cell lines studied. In addition, the data show that upregulation of MCP-1 gene expression is independent of the biochemical nature of the phagocytosed particle. Phagocytosis of latex beads which are commonly regarded as inert and do not induce TNF gene expression,17 upregulate MCP-1 gene expression by monocytic cells to the same extent as phagocytosis of M. tuberculosis. In contrast, phagocytosis of M. tuberculosis but not zymosan or latex upregulated

MCP mRNA ga,“;l;;. Reading)

‘0’ 9o 80 7. 60 50 40 30 20 10 n

Time:

0

1

2

4

8 24

0

1

2

4

8

24

MCP:

Stimulus:

M. tuberculosis

Zymosan

H37-RV Figure 3. MCP-1 gene expression in Mono Mac 6 cells occurs at 4 hours and increases over 24 hours following phagocytosis of zymosan or M. tuberculosis strain H37-Rv. Autoradiographs and their graphed representation following computerized densitometric analysis are shown. p-actin and the equivalent 2% and 18.5 ribosomal RNA bands indicate equal loading of total RNA on gels.

induces

MCP-1

gene expression

I 153

gene expression of IL-& a cytokine chemotactic for T cells,3 in THP-1 cells by 24 hours and in Mono Mac 6 cells within 2-4 hours. The documented increase in cellular IL-8 mRNA levels was accompanied by release of extremely high IL-8 concentrations into culture medium.17 In contrast, we have unpublished observations that phagocytosis of M. tuberculosis resulted in relatively decreased production of IL-6, a cytokine with known anti-mycobacterial activity,4 compared to the effect of other stimuli on human monocytic cells. MCP-1 gene expression appeared to be stimulated by phagocytosis per se and continued for a minimum of 5 days. Since latex particles upregulate MCP-1 gene expression, this phenomenon did not appear to be dependent upon interaction of phagocytosed particles with specific membrane receptors such as those identified as involved in phagocytosis of zymosan.ls Recruitment of monocytes following phagocytosis appears antigen-independent while recruitment of T lymphocytes, a function of IL-8 that has been demonstrated in vitro and in vivo,s may be dependent on the nature of the phagocytosed particle. These observations may partially explain the histological differences between foreign body granuloma comprised principally of cells of the monocyte lineage and tuberculous granuloma which are rich in T lymphocytes as well as monocytes. However, post-transcriptional processing may affect MCP-1 secretion following phagocytosis. It will be possible to study this further both in monocyte cell lines and in alveolar macrophages from human volunteers when ELISAs to measure secreted protein are marketed. The observed increase in MCP-1 gene expression was primarily dependent on the initiation of new transcriptional activity as pre-treatment with actinomycin D prevented MCP-1 mRNA accumulation. Any role for increased MCP-1 mRNA stability cannot be assessed from this study but is clearly not initially the principal reason for increased gene expression.

% Max 100 88 3 28 # Time :

0

1

2

4

8

24

0

1

2

4

8

24

0

0

1

2

4

8

MCP: p-actin: 28118 S mRNA: Stimulus:

M. tuberculosis

H37-RV

M.tuberculoris

H37-RV

M.tuberculosis

+ anti-TNF

H37-RV

24

Figure 4. MCP-I gene expression is similar following phagocytosis of either the virulent H37-Rv strain of M. tuberculosis or the avirulent H37-Ra strain by THP-1 cells. In addition, pre-treatment of cells with 0.1 nM monoclonal antiTNF did not alter MCP-1 gene expression. Autoradiographs and densitometric analysis are shown. rRNA bands are p-actin and equivalent as previously.

154 I Friedland

et al.

CYTOKINE,

The inhibition of protein synthesis by cycloheximide did not prevent MCP-1 gene expression and tended to increase it which may suggest that there may be some constitutive expression of downregulatory proteins. MCP-1 gene expression following phagocytosis was not influenced by the degree of cellular differentiation of the phagocytosing monocytic cell line. MCP-1 mRNA was first detected at 4 hours and increased over 24 hours in both monomyelocytic THP-1 cells and the phenotypically more differentiated macrophage-like Mono Mac 6 cells. This contrasts with IL-8 gene expression following phagocytosis of M. tuberculosis which progressively increases during 24 hours in THP-1 cells but rises to maximal levels after 2-4 h and then declines in Mono Mac 6 cells. In addition, the magnitude of IL-8 mRNA accumulation is greater in THP-1 cells.17 Increased MCP-1 gene expression following phagocytosis may be an important response in all cells of the monocyte lineage. Virulent strains of mycobacteria might potentially interfere with host immune responses, possibly by impairing monocyte recruitment to granuloma and hence become established in vivo, but our findings do not support such a hypothesis. Phagocytosis of the virulent strain of M. tuberculosis resulted in induction of MCP-1 gene expression very similar in magnitude and time course to that which followed phagocytosis of avirulent M. tuberculosis. This was not surprising in view of the other data we have accumulated which has demonstrated the non-specific nature of increased MCP-1 gene expression following phagocytosis. However, this was in contrast to previously reported

Actinomycin Cycloheximide

D

-

-

+

~

-

+

-

+

-

-

+

-

III Stimulus:

Zymosan

H37-Rv M.ruberculosis

Figure 5. unchanged the protein pre-treatment

MCP-1 gene expression following phagocytosis or induced by pre-treatment of THP-1 ceils synthesis inhibitor cycfoheximide but prevented with actinomycin D which inhibits transcription.

is with by

The figure shows autoradiographs and densitometric analysis 16 hours after phagocytosis of zymosan or M. tuberculosis strain H37-Rv either in normal culture medium or in the presence of added cycloheximide or actinomycin D as indicated. rRNA bands are equivalent.

findings

Vol.

5, No. 2 (March

that virulent

1993:

150-156)

but not avirulent forms of prevent induction of IL-l and IL-6 in human monocytesl9 and may thus escape host immune defences. Further experiments we have performed (not shown) indicate that neither IL-l gene expression nor IL-6 secretion is modulated by M. tuberculosis strain virulence in addition to the data on MCP-1 gene expression presented in this paper. Thus the more virulent strain of M. tuberculosis does not appear to downregulate cytokine gene expression or secretion from human monocytes. MCP-1 gene expression appears to be independently regulated following phagocytosis. Although TNF may stimulate MCP-1 gene expression,10 upregulation of this gene in human monocytic cells following phagocytosis was not dependent on autologous secretion of TNF; monoclonal antibody directed against TNF did not decrease the magnitude or alter the kinetics of accumulation of MCP-1 mRNA. TNF is central to granuloma development and treatment with monoclonal anti-TNF reduced granuloma formation and increased mycobacterial multiplication in murine M. bovis infection.2 In addition, TNF restored the ability of SCID mice to form granuloma to Schistosomiasis mansoni eggs.1 However, MCP-1 mRNA accumulation appeared independent of any autologous effect of TNF in human monocytic cells. Although TNF apppears important in development of granuloma, other cytokines including MCP-1 may be more critical in recruitment of those cell types which are subsequently organized into granuloma. Finally, MCP-1 gene expression was not altered by pre-infection of THP-1 cells with HIV. Plasma concentrations of cytokines such as TNF and IL-6 may be elevated in clinical HIV infection20J1 although there are as yet no published data on plasma MCP-1 concentrations in patients. However, monocyte gene expression of these pro-inflammatory cytokines and IL-l or interferon-p following in vitro stimulation was not affected by cellular HIV infection although there was suppression of the Interferon (Y gene.22 Interferon (Y is known to alter HIV release from monocytes23 and prevent development of productive HIV infection as monitored by cytopathologic effects, ~214 core antigen release and detection of viral nucleic acids in cell culture lysates.24 Our observations on MCP-1 gene expression in HIV-infected monocytic cells are consistent with data available for many of cytokines other than interferon-a. However, it is known that in AIDS patients, infection by M. tuberculosis is frequently associated with absent or poorly formed granuloma on histological examination.14 The data presented suggests that although phagocytosis is important in the recruitment of monocytes there are other factors that are critical in development of granuloma. Granuloma formation is likely to be Mycobacterium

avium intracellulare

PhagocytosisinducesMCP-1gene expressioni 155 a complex interaction dependent both on secreted cytokines and intercellular contact. MCP-1 may be pivotal in the initial recruitment of monocytes to sites of subsequent granuloma formation. Finally, the role of the very recently described MCP-2 and MCP-325 in monocyte recruitment following phagocytosis has yet to be determined.

been boiled and washed 7 times in endotoxin-free water served as positive particulate control. Inert latex beads, 3p in diameter (Sigma, Poole, UK), acted as a (negative) particulate control.

Experimental

Protocols

THP-1 or Mono Mac 6 cells were incubated with M. H37-Rv and chemotactic cytokine mRNA accumulation assessedat 0, 1, 2, 4, 8 and 24 h after mixing or at daily intervals over 5 days. Concurrently control experiments using zymosan or latex beads were performed using the same protocol. At each time point, all cells in one PTFE vial were pelleted by centrifugation at 1000 g at 4°C for 5 minutes before RNA extraction (see below). Further data was collected following phagocytosis of the avirulent strain of M. tuberculosis, H37-Ra and compared with data on the virulent strain H37-Rv. Additional experiments were performed following pre-treatment of THP-1 cells with 0.1 nM monoclonal antibody to human TNF (CB0006; Celltech, Berkshire, UK) to investigate the role of this cytokine in MCP-1 gene expression. Experiments were also performed in which THP-1 cells were pre-treated for 30 minutes with either cycloheximide (final cont. 10 pg/ml) or actinomycin D (final cont. 5 pg/ml) which inhibit protein synthesis and transcription respectively. The cells were then exposed to phagocytic stimulus as previously described and MCP-1 gene expression assessedat 16 h. Finally, experiments were performed using HIV-infected THP-1 cells to determine the effect of pre-infection by this virus on chemotactic cytokine gene expression. tuberculosis

MATERIALS AND METHODS Cells and Cell Culture Two established human monocyte cell lines were investigated. THP-1 cells were obtained from American Type Culture Collection (TIB 204) and are an acute myelomonocytic line.26 Mono Mac 6 cells (the kind gift of Dr Ziegler-Heitbrock, University of Munich, Germany), derived from peripheral monoblastic cells, are phenotypically more differentiated and more closely resemble macrophagesz7 Cells were suspended in RPMI-1640 supplemented with 10% fetal calf serum (endotoxin-free), 2 mM glutamine and 1% ampicillin (all Sigma, Poole, UK). Gentamicin was not used as it may inhibit the growth of M. tuberculosis. 107cells were cultured in a 5 ml volume at 37°C in 5% CO in PTFE vials (Pierce, UK) to prevent adherence of monocytes and consequent cytokine secretion.28,29

HIV Infection

of THP-I

Cells

Some THP-1 cells were infected with HIV-l, strain RF30 which was obtained from the Medical Research Council (UK) Aids Directed Programme. 2 x 106 THP-1 cells were incubated in 1 ml of viral supernatant (6 x 104 TCID,,) for 2 h at 37°C in 5% CO,. The THP-1 cells were washed in RPMI-1640, resuspended and cultured for 10 days prior to phagocytosis experiments. HIV infection of these cells was confirmed by measuring culture medium reverse transcriptase activity.31

Phugocytic

Stimuli

Two strains of M. tuberculosis were investigated, the virulent H37-Rv and the less virulent H37-Ra.15 The organisms were initially obtained from the National Collection of Type Cultures (Colindale, UK) and were cultured in Dubos medium enriched with albumen Cohn fraction V and dextrose and containing 200 iu/ml polymixin B, 100 pg/ml carbenicillin, 10 @ml trimethoprin and 10 ug/ml amphotericin B (all Sigma, Poole, UK). Prior to phagocytosis experiments, mycobacteria were transiently sonicated in order to break up clumps and subsequently cultured in RPM1 for the duration of the experiment. Approximately 108 mycobacteria were added to 107 monocytic cells. Previous experiments examining phagocytosis of M. tuberculosis using a modified Kinyoun staining technique showed that approximately 65% of cells (both types) occurred within 30 minutes of mixing mycobacteria and cells. Less than a further 10% of cells ingest mycobacteria over the following 24 h.17 Zymosan granules (Sigma, Poole, UK) which had

RNA Extraction

and Northern

Blotting

Cells, pelleted by centrifugation (1000 g for 5 min at 4”C), were homogenized in 4 M guanidine thiocyanate, 2.5 mM Tris pH 7.0, 0.5% N-lauroylsarcosine and 0.1 M 2-mercaptoethanol. Subsequently RNA underwent double phenol-chloroform and chloroform-isoamyl alcohol extraction, precipitation in isopropanol and then in ethanol before being resuspended in DEPC water. After quantification on an optical densitometer (Pye/Unicam SP6-450) equal amounts of RNA (10-15 pg) were run on denaturing formaldehyde 1% agarose gels, transferred by capillary blotting to Hybond-N (Amersham International, Buckinghamshire, UK) and fixed by exposure to ultraviolet light.

Oligonucleotkie

Probing

Blots were prehybridized with 6 x SSC, 1 x Denhardt’s, 0.5% SDS, 0.05% sodium pyrophosphate, 50 @/ml polyadenylic acid and 100 @ml transfer RNA and then hybridized with y-3zP end-labelled oligonucleotide probes. The MCP-1 probe (the generous gift of Drs RM Strieter and SL Kunkel, University of Michigan Medical School, USA) was a 30 mer,9 IL-8 a 30 mer and B actin a 42 mer.32 Blots were washed and then autoradiographed with intensifying screensat -70°C for 24-48 hours. Between probings, blots were stripped by heating for 1 h at 65°C in a solution of 0.005 M Tris-HCl pH 8.0, 0.002 M Na,EDTA and 0.1 x Denhardt’s.

156 I Friedland

CYTOKINE,

et al.

Autoradiographs were transferred using a Scanjet Plus (Hewlett Packard, Colorado, USA) to a Macintosh IIsi computer (Apple Computer Inc., California, USA) and data quantified with Image 1.41 (from National Institutes of Health Research Services Branch, USA).

Acknowledgements

JSF is a Medical Research Council (UK) Training Fellow. RJS and GEG are supported by the Medical Research Council (UK) and GEG by the Welicome Trust of Great Britain.

REFERENCES 1. Amiri P, Lock&y RM, Parslow TG, Sadick M, Rector E, Ritter D, McKerrow JH (1992) Tumour Necrosis Factor Q restores granulomas and induces parasite egg-laying in schistosome-infected SCID mice. Nature 356:60&607. 2. Kindler V, Sappino A, Grau GE, Piguet P, Vassalli P (1989) The inducing role of Tumour Necrosis Factor in the development of bactericidal granulomas during BCG infection. Cell 56:731-740. 3. Larsen CG, Anderson AO, Appella E, Oppenheim JJ, Matsushima K (1989) The Neutrophil-Activating Protein (NAP-l) is also chemotactic for lymphocytes. Science 243:1464-1466. 4. Flesch IEA, Kaufmann SHE (1990) Stimulation of antibacterial macrophage activities by B-Cell Stimulatory Factor 2 (interleukin-6). Infect Immun 58:269-271. 5. Matsushima K, Larsen CG, DuBois GC, Oppenheim JJ (1989) Purification and characterization of a novel monocyte chemotactic and activating factor produced by a human myelomonocytic cell line. J Exp Med 169:1485-1490. 6. Yoshimura T, Robinson EA, Tanaka S, Appella E, Kuratsu J. Leonard EJ (1989) Purification and amino acid analysis of two human glioma-derived monocyte chemoattractants. J Exp Med 169:1449-1459. 7. Rollins BJ, Stier P, Ernst T, Wong GG (1989) The human homolog of the JE gene encodes a monocyte secretory protein. Mol Cell Biol9:4687-4695. 8. Colotta F, Borre A, Wang JM, Tattanelli M, Maddalena F, Polentarutti N, Peri G, Mantovani A (1992) Expression of a monocyte chemotactic cytokine by human mononuclear phagocytes. J Immunol 148:76@765. 9. Strieter RM, Wiggins R, Phan SH, Wharram BL, Showell HJ, Remick DG, Chensue SW, Kunkel SL (1989) Monocyte Chemotactic Protein gene expression by cytokine-treated human fibroblasts, and endothelial cells. Biochem Biouhvs Res Commun 162:694-700. 10. Larsen CG, Zachariae COC, Oppenheim JJ, Matsushima K (1989) Production of Monocyte Chemotactic and Activating Facror (MCAF) by human dermal fibroblasts in response to Interleukin 1 or Tumor Necrosis Factor. Biochem Bioohvs . , Res Comm 160:140>1408. 11. Standiford TJ, Kunkel SL, Phan SH, Rollins BJ, Strieter RM (1991) Alveolar macrophage-derived cytokines induce Monocyte Chemoattractant Protein-l expression from human pulmonary type II-like epithelial cells. J Biol Chem 266:9912-9918. 12. Rook GAW, Taverne J, Leverton C, Steele J (1987) The role of gamma-interferon, vitamin D3, metabolites and tumour necrosis factor in the pathogenesis of tuberculosis. Immunology &

,

62:229-234.

13. Kochi A (1991) The global tuberculosis situation and the new control strategy of the World Health Organisation. Tubercle 7211-6.

Vol. 5, No. 2 (March 1993: 15@156)

14. Nambuya A, Sewankambo N, Mugerwa J, Goodgame R, Lucas S (1988) Tuberculous lymphadenitis associated with Human Immunodeficiency Virus (HIV) in Uganda. J Clin Pathol41:93-96. 15. Allen BW (1969) Mycobacterium tuberculosis Strain H37Rv. J Med Lab Technol26:389-390. 16. Espevik T, Nissen-Meyer J (1986) A highly sensitive cell line, WEHI 164 clone 13, for measuring cytotoxic factor/Tumor Necrosis Factor from human monocytes. J Immunol Methods 95:9%105. 17. Friedland JS, Remick DG, Shattock R, Griffin GE (1992) Secretion of Interleukin-8 following phagocytosis of Mycobacterium tuberculosis by human monocyte cell lines. Eur J Immunol 22:1373-1378.

18. Czop JK, Kay J (1991) Isolation and characterisation of B-glucan receptors on human mononuclear phagocytes. J Exp Med 173:1511-1520. 19. Michelini-Norris MB, Blanchard DK, Pearson CA, Djeu JY (1992) Differential release of Interleukin (IL)-la, IL-18, and IL-6 from normal human monocytes stimulated with a virulent and an avirulent isogenic variant of Mycobacterium avium-intracellulare complex. J Infect Dis 165:702-709. 20. Lahdevirta J, Maury CPJ, Teppo A, Repo H (1988) Elevated levels of circulating CachectiniTumour Necrosis Factor in patients with Acquired Immunodeficiency Syndrome. Am J Med 85:28!+291. 21. Breen EC, Rezai AR, Nakajima K, Beall GN, Mitsuyashi RT, Hirano T, Kishimoto T, Martinez-Maza 0 (1990) Infection with HIV is associated with elevated IL-6 levels and production. J Immunol144:480-484. 22. Gendelman HE, Friedman RM, Joe S, Baca LM, Turpin JA, Dveksler G, Meltzer MS, Dieffenbach C (1990) A selective defect of Interferon o production in Human Immunodeficiency Virus-infected monocytes. J Exp Med 172:1433-1442. 23. Koyanagi Y, O’Brien WA, Zhao JQ, Golde DW, Gasson JC, Chen ISY (1988) Cytokines alter production of HIV-1 from primary mononuclear phagocytes. Science 241: 167F1675. 24. Kornbluth RS, Oh PS, Munis JR, Cleveland PH, Richman DD (1989) Interferons and bacteria1 lipopolysaccharide protect macrophages from productive infection by Human Immunodeficiency Virus in vitro. J Exp Med 169: 1137-l 151. 25. van Damme J, Proost P, Lenaerts J-P, Opdenakker G (1992) Structural and functional identification of two human, tumor-derived Monocyte Chemotactic Proteins (MCP-2 and MCP3) belonging to the chemokine family. J Exp Med 1765965. 26. Tsuchiya S, Yamahi M, Yamaguchi Y, Kobayashi Y, Konno T, Tada K (1980) Establishment and characterisation of a human acute monocytic leukemia cell line (THP-1). Int J Cancer 26:171-176. 27. Ziegler-Heitbrock HWL, Thiel E, Futterer A, Herzog V, Wirtz A (1988) Establishment of a human cell line (Mono Mac 6) with characteristics of mature monocytes. Int J Cancer 41:456-461. 28. Haskill S, Johnson C, Eierman D, Becker S, Warren K (1988) Adherence induces selective mRNA expression of monocyte mediators and proto-oncogenes. J Immunol 140:169&1694. 29. Kasahara K, Strieter RM, Chensue SW, Standiford TJ, Kunkel SL (1991) Mononuclear cell adherence induces Neutrophil Chemotactic Factor/Interleukin-8 gene expression. J Leuk Biol 50:287-295. 30. Popovic M, Sarngadharan M, Read E, Gallo R (1984) Detection, -isolation and continuous production of cytopathic retrovirnses (HTLV-III) from oatients with AIDS and ore-AIDS. Science 224:497-499. ’ ’ 31. Potts BJ (1990) Mini reverse transcriptase (RT) assay. In Aldovini A, Walker BD (eds). Techniques in HIV research, Stockton Press, New York, pp 103-106. 32. Stricter RM, Kunkel SL, Showell HJ, Remick DG, Phan SH, Ward PA, Marks RM (1989) Endothelial cell gene expression of a Neutrophil Chemotactic Factor by TNF-alpha, LPS and IL-1B. Science 243:1467-1469.