British /ournu) of Haernatology. 1990. 74, 17-23

Secretion of neutrophil secondary granules occurs during granulocyte-macrophage colony stimulating factor induced margination

S . DEVEREUX, J. B. PORTER, K. P. HOYES, R. D. ABEYSINGHE, R. SAIB A N D D. C. LINCH Depnrtment of Haematology, University College and Middlesex Hospital School of Medicine, London

Received 22 December 1988; accepted for publication 30 August 1989

Summary. The effects of recombinant human granulocyte- macrophage colony-stimulating factor (rhGM-CSF) on neutrophil lactoferrin (LF) and transcobalamin (TC) 1 and 3 secretion were determined in vitro and during in vivo administration in humans. In whole blood, in vitro incuba- tion with GM-CSF reproducibly produced a rise in plasma LP concentration ( P < 0 . 0 5 ) whereas in purified neutrophils the results were variable. Exposure of whole blood to GM-CSP also resulted in a significant rise in plasma TC 1 and 3 (190&60%. P<0.05) . The response was dose dependent

with maximal effect at GM-CSF concentrations of 10 ng/ml and above. rhGM-CSF was administered on seven occasions to six patients with malignant disease prior to chemotherapy. Plasma LF and unsaturated TC 1 and 3 levels rose signifi- cantly in each patient studied and the rise coincided with the initial neutropenia due to margination that occurs during infusions of rhGM-CSF. Patients receiving rhGM-CSF may therefore have hypofunctional neutrophils due to secondary granule depletion.

The adherence of neutrophils to the vascular endothelium and subsequent egress into the tissues is of central impor- tance in the inflammatory response. At any one time approximately half of the intravascular pool of neutrophils is marginated or adherent to the endothelium. Stimuli such as adrenaline or glucocorticoids cause neutrophils to leave the marginated pool and enter the circulation accounting for the rapid neutrophilia induced by these agents (Athens et ul, 1961). Other factors such as endotoxin (Craddock et nl, 1960) or the complement fragments C3a and C5a (Craddock et al, 1977) cause a rapid shift into the marginated compart- ment and consequent neutropenia. It is likely that one of the main mechanisms by which locally produced complement fragments and other inflammatory mediators encourage the localization of neutrophils and monocytes in areas of inflam- mation is by increasing their adherence to endothelial cells.

A family of related adherence promoting glycoproteins, expressed in many cell types, has been characterized (Springer et al, 1987). Both neutrophils and monocytes express a heterodimeric cellular adhesion molecule (CAM) whose alpha chain is recognized by monoclonal antibodies of

Correspondence: Professor D. C. Linch, Department of Haematology . University College and Middlesex Hospital School of Medicine. London WC1 E6HX.

the C D l l series and the beta chain by anti CD18 antibodies. A receptor called ICAM-1 has been identified as the ligand for lymphocyte function antigen1 (LFA-1, CD1 la/CD18) depen- dent leucocyte adherence (Makgoba et d, 1988). Adhesion of neutrophik to the vascular endothelium is thought to depend at least in part on the binding of CAMs to this endothelial ligand (Belvilacqua et al, 198 7). The nature of other receptors for the CAMs has not yet been defined.

A proportion of adherent neutrophils cross the endothelial barrier and enter the tissues. What controls this and how it occurs are not well understood. Neutrophils in pathological and experimentally induced inflammatory exudates have reduced secondary granule content (Wright & Gallin, 1979) and it has been suggested that secondary granule products play some part in this process possibly by digesting a path through interendothelial junctional and basement mem- brane (Harlan, 1985).

The haemopoietin granulocyte-macrophage colony- stimulating factor (GM-CSF) has been shown to increase neutrophil CAM expression in vitro (Arnaout et uZ, 1985). We have shown that recombinant human GM-CSF (rhGM-CSF) causes transient margination of neutrophils and monocytes mainly within the lungs (Devereux et al, 1987). This is associated with an in vivo increase in neutrophil CAM expression (Socinski et 01, 1988; Devereux et nl, 1989). In the

1 7

18 S. Devereux et a1 resting state, neutrophil CAMS are mainly located in the secondary granule membrane but following exposure to chemotactic factors, or other stimuli, are translocated to the surface membrane where they mediate adherence to endo- thelial cells. The extracellular ligand binding portion of the molecule is thought to be oriented to the interior of the specific granule (Todd et al, 1984; Bainton et al, 1987). We have therefore studied the effects of rhGM-CSF on neutrophil secondary granule release in vitro and in vivo to determine whether CAM upregulation is associated with secretion of granule contents.

PATIENTS AND METHODS

Reagents Yeast derived rhGM-CSF was supplied by Immunex (Seattle, Wash., U.S.A.) and E. coli derived rhGM-CSF by Behr- ingwerke (Marburg, F.D.R.). The endotoxin contamination was less than 0.4 ng/mg and 0.3 ng/mg respectively. All in vitro studies were performed with yeast derived product. The chemotactic peptide f-methionyl-leucyl-phenylalanine (fmlp) was obtained from Sigma (U.K.). Medium alone was used as a negative control agon i~ t .~~Co vitamin Blz stock solution was prepared by diluting 0.66 ,itg/ml BIZ with an activity of 370 kBq/ml (Amersham International) 1 : 100 in 0.9% saline. This was further diluted 1 : 3 on the day of assay to yield a solution containing 2.2 ng/ml B12 with an activity of 1.2 kBq/ml. Norit A charcoal (Sigma) was made up to a 5% solution in distilled water and mixed 1 : 1 with a 1% solution of bovine serum albumin (Sigma V BSA) on the day of assay. Quso G32 microfine precipitated silica (Philadelphia Mining Co.) was used dry and measured using a specially constructed scoop holding approximately 10 mg.

In-vitro studies Secretion studies performed on purijied neutrophils. Venous

blood was taken into acid citrate dextrose solution (ACD-A) and red cells removed by sedimentation with 1% hetastarch for 45 min at room temperature. The supernatant was layered onto Ficol-Hypaque and centrifuged at 300 g for 25 min at 25°C. The plasma, interface and gradient were discarded and contaminating red cells removed by hypotonic lysis at 4°C. The resulting purified granulocytes were washed twice in RPMI (Gibco) and resuspended at a concentration equal to that in whole blood. Aliquots of purified granulo- cytes were incubated with varying doses of agonist for periods of up to 1 h and then rapidly cooled to 4°C. Each experiment was performed in triplicate. The cells were pelleted by gentle centrifugation and the supernatants harvested and stored at - 70°C until assayed for LF.

Secretion studiesperformed on whole blood. Venous blood was taken from healthy laboratory personnel into preservative- free heparin 20 u/ml at room temperature. Aliquots were warmed to 37°C and incubated with agonists at varying doses for periods of up to 1 h. Each experiment was performed in triplicate. At the end of this period, the samples were rapidly cooled to 4OC and plasma separated at 4°C by centrifugation for 10 min at 1000 g. Care was taken to avoid disturbing the cellular pellet and the plasma samples were stored at - 20°C until needed for assay.

In-vivo studies Putients. Studies were performed on two groups of patients.

Local ethical committee approval and informed written consent from each patient were obtained. One group of three patients with advanced solid tumours not responding to conventional therapy received 10 d treatment with rhGM- CSF as part of a phase 1 study. The second group of three patients with resistant Hodgkin's disease received rhGM-CSF as an adjunct to high-dose chemotherapy and autologous bone marrow transplantation. The effects of rhGM-CSF in this group were determined prior to chemotherapy. Yeast derived rhGM-CSF was administered to three patients at a dose of 15 pg/m2/h and E. coli derived rhGM-CSF in an equimolar dose on four occasions in three patients. Infusions were continued for up to 3 h.

Venous blood samples were taken through an indwelling venous cannula into sodium ethylenediaminetetraacetate (EDTA) 1 mg/ml on ice, before and at intervals during infusion of rhGM-CSF. Samples were not taken after GM-CSF administration had ceased. Plasma was separated within 15 min by centrifugation at 1000 g for 10 min at 4°C. Care was taken not to disturb the cellular pellet during separation and aliquots of plasma were frozen and stored at -20°C.

Assay methods Unsnturated TCl + 3 assay. Total unsaturated BIZ binding

capacity and then plasma transcobalamin 1 + 3 (TC1+ 3) were assayed by a modification of the method of Jacob et a1 (1977). 1.8 ml microfuge tubes were labelled as Standard (A), Blank (B) and Test (C). To tubes A and B, 500 pl and 100 pl of 0.9% saline were added and 100 p1 of test sample was added to tube C. 200 ,it1 of diluted j7Co B12 was added to each tube, mixed and incubated for 30 min at room temperature. 400 ,it1 of albumin coated charcoal was then added to tubes B and C. After a further 10 min incubation the samples were centrifuged at high speed in an MSE microfuge and the supernatant containing total Bl2 binding protein saturated with 57Co B12 transferred to another tube for counting or further separation into TC1+ 3 and TC2.

The glycoproteins TC1+ 3 were separated from the poly- peptide TC2 by adsorption to microfilm precipitated silica. Each supernatant obtained from the total Blz binding assay was mixed and incubated with 20 mg Quso G32 for 10 min at room temperature. Samples were then centrifuged at high speed in an MSE microfuge for 10 min and the supernatants removed for counting in a gamma counter. Following subtraction of background, Bl? binding capacity, was calcu- lated for each sample using the formula: Biz binding capacity (ng/l) = Test C (cpm) -Blank B (cpm)/Standard A (cpm) x 2200.

LF assay. Samples were assayed for LF concentrations using an enzyme linked immunoassay (ELISA) system incor- porating the use of the labelled avidin-biotin technique (Guesdon et al, 1979). 50 pl aliquots of prediluted samples or standards were incubated in 96 well ELISA plates (Linbro- Titerteck) previously coated with polyclonal anti-LF (Dako). Following four washes in 0.1% Tween/phosphate buffered saline, biotinylated polyclonal anti-LF was added to each well for 1 h at 20°C. After four further washes, 50 p1 of avidin

Secretion of Neutrophil Secondary Granules 19

peroxidase conjugate (Miles) in borate-saline buffer pH 8.4 was added to each well for 1 h at 20OC. Unbound conjugate was removed by a further four washes and a substrate buffer containing 34 mg of o-phenylenediamine in 100 ml of citrate phosphate buffer with 50 p1 HzOz was added. The reaction was stopped with 100 ~ 1 1 2 . 5 % H2SO4 and the optical density read in an automated EISA plate reader at 492 nm.

Statistical methods The differences between unstimulated and stimulated secre- tion were compared using Student’s t-test for paired data. In

Table I. GM-CSF induced lactoferrin release from neutrophils

Whole blood lactoferrin concentration (ng/ml) ferrin concentration (ng/ml)

EXP. NO GM-CSF GM-CSF NO GM-CSF GM-CSF

Isolated neutrophil lacto-

1 422.1 678.5 38.5 29.0 2 553.3 610.7 246.5 283.4 3 511.4 616.0 261.3 333.2 4 459.1 503.1 165.0 170.2

Mean 481.5 602.1 177.8 203.9 SEM 25.2 36.4 56.0 67.5

The changes in plasma (whole blood) or supernatant (isolated neutrophils) lactoferrin concentration (ngiml), following a 1 h incubation with rh-GM-CSF 10 mg/ml or diluent control. In each experiment on whole blood, GM-CSF produced a highly significant rise in lactoferrin (P<0,001). The mean value of lactoferrin concentration for the four experiments was also increased in the GM- CSF treated whole blood samples (Pc 0.05 ).

100

80

= 60 01 C

. - 0 + .- 2 40

20

0

l a

RPMl GMO lnglrnl GM lOnglml GM lpglml

St imulus

in-vivo studies, differences in plasma levels between time zero values and later time points were again compared using Student’s t-test for paired data.

RESULTS

In-vitro studies In initial experiments the release of lactoferrin from whole blood and purified neutrophils in response to incubation with GM-CSF 10 ng/ml was assessed (Table 1). In whole blood a reproducible rise in lactoferrin concentration was observed. In each experiment, nine data points were obtained for each value shown and the difference between the GM-CSF treated and control samples was highly significant in each case ( P < O . O O l ) . Following a 1 h incubation with or without rhGM-CSF, the neutrophil viability was > 98% as determined by ethidium bromide exclusion. In the purified neutrophils, the results were more variable and the means of the GM-CSF treated and control samples were not significantly different. Further experiments were therefore performed on whole blood.

Exposure to rhGM-CSF resulted in a significant ( P > 0.0 5) increase in unsaturated TCl + 3 release above that induced by medium alone (195% f 60, n = 4, P < 0.05). The response to rhGM-CSF was dose dependent with a maximal effect observed at a rhGM-CSF concentration of 10 ng/ml (Figs l a and b). The effects of rhGM-CSF were progressive with time (Fig 2). There was little release of TC1+3 within the first 5 min contrasting with the rapid early rise induced by fmlp. Similar results were obtained for LF release: the LF concen- tration in the supernatant ( f standard error) after 0, 30 and 60 min being 3 3 3 f 4 8 , 483 .4k2 .5 and 6 7 8 . 5 f 4 5 ng/ml respectively.

800

600

- - E . 01 C - C ‘ z 400

a, 0 0 m _1

L

L -

200

0

I b r-

RPMl GMO 1ng;ml GM 10ngInil CM l u r : II

St imulus

Fig 1. Release of unsaturated TCI + 3 (Fig l a ) and LF (Fig 1 b) from whole blood at 1 h in response to diluent and a varying dose of rhGM-CSF (GM). Each assay was performed in triplicate and results expressed as the mean +standard error of the mean (SEM).

20 S. Devereux et a1 In-vivo studies The plasma levels of unsaturated TC1+3 and LF were measured serially on six and seven occasions before and during infusions of rhGM-CSF given prior to chemotherapy. 1 h after commencing the infusion of rhGM-CSF, the un- saturated TCl + 3 level had risen in all of six patients (P<0.01) and the plasma LF in six out of seven patients

100

80

- . err S

0 60 + 0 I-

7

40

20

- GM -CSF lOngiml - fmlp10 M - R P M l

-7

- GM -CSF lOngiml - fmlp10 M - R P M l

-7

I ' ~ , ' ~ I * ' I

1 5 3 0 4 5 6 0

Time ( m i n u t e s )

Fig 2. Release of unsaturated TC1+ 3 from whole blood in response to M f-methionyl-leucyl-phenylalanine (fmlp). 10 ng/ml rhGM- CSF and diluent for varying periods. Each assay was performed in triplicate and the results expressed as the mean *standard error of the mean (SEM).

( P < 0 . 0 5 ) (Table 11). Yeast-derived rhGM-CSF was used in three patient studies and E. coli derived material in the other four. The responses were similar with both preparations (Table 11).

Three patients received rhGM-CSF infusions lasting for 3 h. The plasma unsaturated TC1+ 3 and LF levels had risen by 30 min and did not significantly rise further after 1 h (Figs 3a and b). The rise in plasma LF and unsaturated TC 1 + 3 level coincided with the onset of neutropenia due to margination seen at the start of rhGM-CSF administration but persisted after demargination. The magnitude of the maximum rise in plasma unsaturated TC1+ 3 and LF level did not correlate with the maximum fall in the neutrophil count.

DISCUSSION

These results clearly show that rhGM-CSF causes neutrophils to release the secondary granule products LF and TC 1 + 3 both in vitro and during in-vivo administration in humans. In-vitro secretion of secondary granule products is, however, sensitive to purification procedures. Reproducible release occurs from unfractionated neutrophils but not from purified cells. Lopez et al (1986) had previously failed to show secondary granule release induced by GM-CSF but had only studied purified neutrophils. This could be because the purification damages the neutrophils and causes excessive release in unstimulated cells, or alternatively impairs the secretory response to GM-CSF by causing release of a labile secondary granule pool upon which rhGM-CSF acts. Fearon & Collins (1983) have previously shown that isolation procedures can cause upregulation of neutrophil CAM expression from storage sites in the secondary granule membrane (Bainton et al, 1987).

The orientation of neutrophil CAMS in the secondary granule membrane, with the extracellular ligand binding domain to the granule interior, is such that release of granule contents would be expected to occur during this process. The fact that the plasma levels of LF are higher than those in the supernatants of purified neutrophils is at least in part due to the presence of LF in normal plasma in vivo.

Table 11. The changes in plasma unsaturated TC1+ 3 and LF levels observed during infusions ofrhCM-CSF at a dose of 15 pg/m2/h on seven occasions in six patients

Patient 1. Patient 2, Patient 3, Patient 4, Patient 5. Patient 6a. Patient hb. Type rhGM-CSF yeast yeast yeast E. coli E. coli E. coli E.coli

LF (nsiml) 0 min

30 min 60 min

120 min 180 min

TC1+ 3 (ng/l) 0 min

30 min 60 min

120 min 180 min

235 517 72 3 930

4 7 73 95 74

238

326 351 49 5

170

200 153 201

95 123 131 145 141

85 8 7

115 137 154

123 115 189 186

51 103 149 130

553 682 667 652

107 171 201 133

264 487 502 532 649

48 I42 172 146

98

Secretiori of Neutrophil Secondary Granules 2 1

300

m c

2 2 N

.- E - zoo 0 - 8

100 0

3a

1 T C

6 0 120 180

Time (minutes)

I

Time (minutes)

Pig 3.Plasma levels of unsaturated TCI + 3 (Fig 3a) and LF (Fig 3b) during 3 h infusions of rhGM-CSF 1 5 pg/mz/h in three patients. Results are expressed as a percentage of the initial value of plasma unsaturated TC1+ 3 or LF and expressed as the mean f SEM. The circulating neutrophil count at the time of unsaturated TCI + 3 and LF assay is also shown and expressed as the mean +SEM.

It is difficult to ascertain whether GM-CSF induced de- granulation is a direct effect or priming of a degranulation response induced by cell manipulation in vitro. The whole blood experiments minimize such artefacts but do not completely exclude such events. It was particularly interest- ing therefore that GM-CSF also causes a rise in LF and TCl + 3 in vivo.

The in-vivo secretion of TC 1 + 3 and LF induced by rhGM- CSF coincides with the initial neutropenia, margination and CAM upregulation seen during the first 2 h of rhGM-CSF treatment (Devereux et al, 1987, 1989; Socinski et al, 1988). All of these phenomena could be explained on the basis of secondary granule exocytosis. The pre-rhGM-CSF adminis- tration levels of plasma LF and unsaturated TC1+3 were broadly within agreement of previously published values (Porter & Burke, 1989; Scott et al, 1974: Jacob et aI, 1977). As expected these values are lower than those obtained from serum samples which lead to variably and unreproducibly raised values (Scott et al, 1974: Bennet & Kokocinski, 1978). It is unlikely that endotoxin contamination is responsible for the in-vitro and the in-vivo findings as the amount of endotoxin present was very low and endotoxin administra- tion results in transient lymphopenia as well as neutropania (Craddock et al, 1960), and no changes in lymphocyte counts were observed in these studies.

We have not formally excluded the possibility that the rise in LF and TC1f 3 is due to decreased clearance rather than to neutrophil secretion. This we believe is unlikely as LF and TCI i- 3 are cleared by different mechanisms and changes in clearance could not account for the in-vitro data.

We have previously shown that the recovery from the rapid neutropenia associated with the commencement of rhGM-CSF infusions is due to re-entry of marginated cells into

the intravascular pool (Devereux et al, 1989). As these cells appear to have released at least a proportion of secondary granules they will for at least a period of time be deficient in secondary granules and may therefore be functionally impaired.

The precise biological role of secondary granules is unclear. In addition to membrane associated adhesion receptors they contain molecules with bacteriostatic functions such as LF and TCs and enzymes some of which may be involved in the migration of neutrophils through the interendothelial junc- tions and basement membrane (Harlan, 1985). Patients with congenital secondary granule deficiency show reduced migration of neutrophils into inflammatory foci and skin windows (Gallin et al, 1982) and neutrophils in experimental and pathological exudates are deficient in secondary granules (Wright & Gallin. 1979). Secondary granules may therefore be required, and lost during migration from the circulation into the tissues.

Preliminary evidence suggests that rhGM-CSF infusions in man cause decreased migration into skin window mem- branes (Addison et al, 1989). GM-CSF has effects on neutro- phi1 mobility in vitro, some studies reporting inhibition of migration (Gasson et al, 1984) and others that it is a chemotactic factor (Wang et al, 1987). GM-CSF is synthesized by activated T lymphocytes, monocytes, endothelial cells and fibroblasts (Clark & Kamen, 1987) and is produced in inflammatory foci (Koury et al, 1983). It is thus likely that locally produced GM-CSF plays a role in the recruitment of neutrophils at sites of inflammation and that the non- directional stimulus of intravenously administered rhGM- CSF might impair subsequent responses to appropriate signals. It seems probable that secondary granule exocytosis and CAM upregulation is one of the mechanisms by which

22 S. Devereux et a1

GM-CSF affects neutrophil localization and that inappro- priate release of these granules by systemic rhGM-CSF might contribute to the impaired migration into skin windows.

Clinical studies with rhGM-CSF have so far focused on its stimulatory effects on the bone marrow with the hope that any increase in circulating neutrophil numbers would result in improved host defence against infection. It is now clear from a number of such studies that rhGM-CSF does indeed have potent effects on myelopoiesis in-vivo. The leucopenia seen in the acquired immune deficiency syndrome and in patients with myelodysplasia is improved (Groopman et aI, 1987; Vadhan-Raj et al, 1987) and the period of neutropenia following both standard chemotherapy and high-dose chemotherapy with autologous bone marrow transplan- tation shortened (Antman rt nl, 1987 ; Brandt et al, 1988: Devereux et al, 1988).

Definitive proof of clinical efficacy in preventing infections, is, however, still lacking and our studies draw attention to the possibility that rhGM-CSF might lead to impairment of certain phagocyte functions in vivo. It is possible that a dose rate of less than the 15 pg/m2/h used in this study might stimulate the bone marrow without causing secondary granule release. Further studies are needed to define the optimum dose and to further examine the effects of rhGM-CSF on phagocyte function and host defence in vivo.

ACKNOWLEDGMENTS

This work was supported by the Kay Kendall Leukaemia Fund and a grant from North East Thames Regional Health Authority. rhGM-CSF was supplied by Immunex Corpo- ration, Seattle, Washington, U.S.A., and Behringwerke, Marburg, F.D.R.

REFERENCES

Addison, I.E., Johnson, B.. Devereux, S., Goldstone. A.H. & Linch, D.C. (1 989) Granulocyte-macrophage colony stimulating factor may inhibit neutrophil migration in-vivo. Clinical and Experimental Immunology. 76, 149-153.

Antman, K.. Griffin, J., Elias, A., Socinski. M.. Whittey, M., Ryan, L., Cannistra, S.. Oette. D.. Frei, E. & Schnipper. L. (1987) Use of rGM- CSF to ameliorate chemotherapy induced myelosuppression in sarcoma patients. Blood 70, (Suppl.), 373.

Arnaout, M.A., Wang, E.A.. Clark, S.C. & Sieff. C.A. (1986) Human recombinant granulocyte-macrophage colony-stimulating factor increases cell to cell adhesion and surface expression of adhesion promoting glycoproteins on mature granulocytes. Journal of Clinical Investigation, 78, 597-601.

Athens, J.W., Haab, O.P.. Raab, S.O., Mauer, A.M., Asembrucker, H.. Cartwright, G.E. & Wintrobe, M.N. (1961) Leukokinetic studies. IV. The total blood circulating and marginal granulocyte pools and the granulocyte turnover rate in normal subjects. Jounal of Clinical Investigation, 40, 989-993.

Bainton, D.F., Miller, L.J.. Kishimoto. T.K. & Springer, T.A. (1987) Leucocyte adhesion receptors are stored in peroxidase negative granules of human neutrophils. Journal of Experimental Medicine. 166,1641-1653.

Bennet, R.M. & Kokocinski, T. (1978) Lactoferrin content of peripheral blood cells. British Journal of Haernatology, 5 1,

Bevilacqua, M.P.. Pober, J.S.. Mendrick, D.L., Cotran, R.S. & Gim- brone, M.A., Jr (1987) Identification of an inducible endothelial- leukocyte adhesion molecule. Proceedings of the National Academy of Sciences of the United States of America, 84, 9238-9242.

Brandt, S.].. Peters, W.P.. Atwater, S.K., Kurtzberg, J.. Barowitz, M.].. Jones, R.B.. Shpall, E.J.. Bast, R.C., Jr, Gilbert, C.J. & Oette, D.H. (1 988) Effect of recombinant human granulocyte-macrophage colony-stimulating factor on hemopoietic reconstitution following high dose chemotherapy and autologous bone marrow transplan- tation. New England Journal ojMedicine. 318, 869-875.

Clark, S.C. & Karnen. T. (1987) The human hematopoietic colony stimulating factors. Science, 236, 1229-1237.

Craddock. C.G.. Perry, S.. Ventzke, L.E. & Lawrence, J.S. (1960) Evaluation of marrow granulocytic reserves in normal and disease states. Blood, 15, 840-855.

Craddock, P.R.. Fehr, J.. Dalmasso, A.P., Brigham. K.L. & Jacob, H.S. (1 9 77) Hemodialysis leukopenia. Journal of Clinical Investigation.

Devereux, S., Bull, H.A., Campos-Costa. D.. Saib, R. & Linch. D.C. (1 989) Granulocyte macrophage colony stimulating factor induced changes in cellular adhesion molecule expression and adhesion to endothelium: in-vitro and in-vivo studies in man. British Journal of Haematology, 71, 323-330.

Devereux, S., Linch, D.C., Campos-Costa, D.. Spittle, M.F. & Ielliffe, A.M. (1987) Transient leucopenia induced by granulocyte macro- phage colony stimulating factor. (Letter). Lancet. ii, 1523.

Devereux,S..Patterson,K.P..Goldstone.A.H.G.&Linch.D.C. (1988) The use of recombinant human granulocyte macrophage colony stimulating factor to accelerate haemopoietic recovery after autologous bone marrow transplantation (Abstract). British Jour- nal of Haematology, 69, 103.

Fearon, D.T. & Collins, L.A. (1983) Increased expression of t 3 b receptor on polymorphonuclear leukocytes induced by chemotac- tic factor and by purification procedures. Journal of Immunology,

Gallin,]., Fletcher, M.. Seligmann. B., Hoffstein, S. &Cehrs, K. (1982) Human neutrophil specific granule deficiency: A model to assess the role of neutrophil specific granules in the evolution of the inflammatory response. Brood. 59, 131 7.

Gasson, J.C., Weisdorf, R.H., Kaufman, S.E., Clark, S.C., Hewick, R.M., Wong. G.G. & Golde. D.W. (1984) Purified human granulo- cyte-macrophage colony-stimulating factor: direct action on neutrophils. Science, 226, 1339-1342.

Groopman, J.E.. Mitsuyasu. R.T.. DeLeo, M.].. Oette. D.H. & Golde, D.W. (1 98 7) Effect of granulocyte-macrophage colony stimulating factor on myelopoiesis in the acquired immune deficiency syn- drome. New England Journal of Medicine, 317, 593-598.

Guesdon. J., Teernynk. T. & Avrameus, S. (1979) The use of avidin- biotin interaction in immunoenzymatic techniques. Journal of Histochemistry and Cytochemistry, 27, 1131-1139.

Harlan, J.M. (1985) Leukocyte-endothelial interactions. Blood, 65,

Jacob, E., Wong. K.T.]. &Herbert, V. (1977) A simple method for the separate measurement of TCs 1 , 2 and 3: normal ranges in serum and plasma in men and women. Journal of Laboratory and Clinical Medicine, 89, 1145-1152.

Koury, M.J., Balmain, A. & Pragnell. I.B. (1983) Induction of granulocyte-macrophage colony stimulating activity in mouse skin by inflammatory agents and tumour promoters. EMBO Journal. 2, 1877-1882.

Lopez, A.F., Williamson, I., Gamble, J.R., Begley, C.G.. Harlan. J.M.4 Klebanoff, S.J., Waltersdorph. A., Wong, G. & Clark, S.C. (1986)

509-52 1.

59,879-888.

130, 370-375.

513-525.

Recombinant human granulocyte-macrophage stimulating factor stimulates in vitro mature human neutrophil and eosinophil function, surface expression and survival. Journal of Clinical Investigation. 78, 12 20- 1 22 8.

Makgoba, M.W., Sanders, M.E., Ginther Luce, G.E.. Dunstin, M.L., Springer, T.A.. Clark, E.A.. Mannuni. P.S. (1988) ICAM 1 a ligand for LFA 1 dependent adhesion of B. T and myeloid cells. Nature, 331, 86-88.

Porter. J.B. & Burke. L.C. (1989) Plasma LF as a marker of granulocyte regeneration following myelosuppressive therapy. (Submitted).

Scott, J.M.. Bloomfield, F.J., Stebbuns, J. & Herbert, V. (1974) Studies on derivation of transcobalamin I11 from granulocytes. Journal of Clinical Investigation. 53, 228-239.

Socinski. M.A.. Canistra, S.A.. Sulivan. R.. Elias. A., Antman, K . , Schnipper. L. & Griffin, J.D. (1988) Granulocyte-macrophage colony-stimulating factor induces the expression of the CD11 b surface adhesion molecule on human granulocytes in-vivo. Blood.

Springer.T.A.,Dustin,M.L..Kishimoto,T.K.&Marlin,S.D.A. (1987) 72, 691-697.

Secretion of Neutrophil Secondary Granules 2 3 The lymphocyte function associated LFA-1, CDZ and LFA-3 molecules: cell adhesion receptors of the immune system. Annual Reviews of lmniimology, 5, 223-252.

Todd. R.F., Arnaout, M.A., Rosin, R.E., Crowley, C.A.. Peters, W.A. & Babior, B.M. (1984) Subcellular localisation of the large subunit of Mol (Mola formerly GP110) a surface glycoprotein associated with neutrophil adherence. Journal of Clinical Investigation, 74,

Vadhan-Raj. S., Keating. M.. LeMaistre, A., Hitelman, W.N., McCre- die. K., Trujillo, J.M.. Bruxmeyer, H.E.. Henney. C. & Gutterman, J.J. (1987) Effects of recombinant human granulocyte-macro- phage colony-stimulating factor in patients with myelodysplastic syndromes. New England Journal of Medicine. 317, 1545-1551.

Wang, J.M.. Colella, S., Allavena, P. & Mantovani. A. (1987) Chemotactic activity of human recombinant granulocyte-macro- phage colony-stimulating factor. Immunology. 60, 439.

Wright, D.G. & Gallin, J.I. (1979) Secretory responses of human neutrophils: Exocytosis of specific (secondary) granules by human neutrophils during adherence in-vitro and exudation in-vivo. Journal of Immunology, 123, 285-294.

1280-1290.