Journal of Immunological Methods, 126 (1990) 7-11

7

Elsevier JIM 05412

Quantification of radiolabelled granulocyte migration in vivo A.J. Pinching 1,,, A.M. Peters 2 and S.H. Saverymuttu 1 , , , Departments of I Medicine and 2 Diagnostic Radiology, Hammersmith Hospital, London, U.K.

(Received 1 June 1989, revised received 7 August 1989, accepted 8 September 1989)

Quantification of the tissue localisation of granulocytes is difficult, particularly in man. With the aim of facilitating such quantification, a technique is described which combines the micropore filter skin window technique with the infusion of autologous rain-labelled granulocytes. The radioactivity in the removable micropore filters, placed on small forearm skin window abrasions, was compared with peripheral blood radioactivity following injection of 1lain-labelled 'pure' granulocytes in normal subjects. Radioactivity in filters from abrasions which were made 8 h or more before injection of labelled cells followed the same time course as the cell-associated radioactivity in whole blood, but radioactivity in filters from abrasions made nearer to the time of labelled granulocyte injection increased to reach a peak 3-8 h after cell injection, at a time when cell-associated blood radioactivity had fallen to about 50% of its initial value. Exercise appeared to induce a transient decrease in radiolabelled granulocyte migration into the filters. This technique offers a means of studying the kinetics of granulocyte migration in vivo. Key words: lllln-labeled granulocyte; Skin window; Inflammation

Introduction It has proved difficult to quantify the tissue localisation of granulocytes in man. It would be useful to be able to do this; for instance different disease states could be compared, and dose-response curves for the defective tissue granulocyte localisation seen during corticosteroid therapy could be established, In this paper, we describe an approach to such quantification by combining the micropore filter

Correspondence to: A.M. Peters, Department of Diagnostic Radiology, Hammersmith Hospital, London, U.K. * Present address: Department of Immunology, St. Mary's Hospital Medical School, London W2 1PG, U.K. * * Present address: Medicine II, St. Georges Hospital Medical

School, LondonSWl7 ORE,U.K.

skin window technique (Addison et al., 1982) with the autologous infusion of ~HIn-labelled pure granulocytes (Peters et al., 1983). The abrasion made for the skin window technique and the overlying filter become a focus of inflammation into which the radiolabelled granulocytes migrate. The filters can be periodically replaced and conveniently transferred to well-counting tubes for radioactivity measurement. This can be compared with radioactivity in blood, thereby extending quantification of the granulocyte migration into the abrasion. By varying the time interval between making the abrasion and injecting the cells, we have studied the rate at which the 'abrasion' matures, i.e., reaches a steady state of radiolabelled granulocyte accumulation. We have also demonstrated how one physiological variable, exercise, can modify radiolabelled granulocyte localisation in the abrasion, thereby

0022-1759/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)

8

illustrating the sensitivity of this combined approach.

Materials and methods Four normal males volunteered for this study, Abrasions were made on the forearms using a dental drill with burr attachment according to Addison et al. (1982). Micropore filters of thickness 3 /~m were placed on the abrasions with a 0.22 g m filter above, the whole being kept moist with Whatman filter paper soaked in R P M I medium and kept stable with parafilm, gauze and tubigrip. Filters were changed at intervals, hourly at first, following injection of the labelled cells, The filters were transferred to well-counting tubes for gamma counting in a Packard 5360 automatic counter, using both lUIn photopeaks. When filters were left in place for periods longer than 1 h, the total radioactivity was divided by the time interval and expressed as an hourly value at the mid-point of the interval, Autologous granulocytes from 60 ml of venous blood, anticoagulated with A C D (NIH, formula A), were labelled with m I n - t r o p o l o n a t e in plasma following separation on a discontinuous Metrizamide-plasma gradient as previously described (Peters et al., 1983; Saverymuttu et al., 1983). The granulocytes were labelled with 2.5 MBq (90 gCi) of r a i n and injected intravenously. Dynamic g a m m a camera ( I G E 400T) imaging at a frame rate of 1 / m i n was performed for 40 min after injection and the data stored on an M D S A2 computer. Regions of interest were drawn over the right lung, liver and spleen for the generation of radioactivity-time curves. Blood samples of 5 ml were taken into E D T A at carefully timed intervals after injection. Cell-free plasma was separated immediately and 1 ml aliquots later counted in the well counter with 1 ml aliquots of whole blood in order to construct a blood disappearance curve of cell-associated radioactivity. Radioactivity in the filters was counted at the same time. In two subjects the effect of exercise on filter radiolabeUed granulocyte localisation was studied, The exercise consisted of running up and down stairs for 15 min.

Results There was no significant delay in the transit of the labelled cells through the pulmonary vasculature following their injection, and the radioactivity-time curves over the liver and spleen showed an early distribution compatible with optimal functional cellular integrity (Saverymuttu et al., 1983). The time courses of peripheral blood (cellbound) and filter ~11In activities are shown in Fig. 1 for a volunteer in whom one abrasion was made 20 h before injection of labelled granulocytes. Both blood and filter time activity curves showed an essentially monoexponential decrease and, apart from the first 2 h, were almost superimposed with half-times of 4.9 h and 4.7 h respectively. When the abrasion was made only 3 h before cell injection, a different time course of filter-associated radioactivity was obtained. Thus, filter radioactivity peaked at 9 h after cell injection, at a time when the circulating radioactivity had fallen to about 50% of the initial value (Figs. 2 and 3). In order to clarify the ' m a t u r a t i o n ' time of the abrasion, three abrasions were made in a single subject at 20, 8 and 4 h before radiolabelled granulocyte injection. Although the 20-h-old abrasion accumulated more radiolabelled granulocytes

l o o - ~ ~'~..,~ ~ ,

a---btoo~ C~e,-bo~.d~ .f i l t e b~ood r s ' cplosma~ "

~ lo 0_ ~

, ~

1

o

lo 2'o 3'0 Time from injection (h) Fig. 1. Time course of rain activity in peripheral blood cells (D), peripheral blood plasma (~), and micropore filters (11)in a normal subject in whom an abrasion had been made 20 h before injection of l]lln-labelled granulocytes. Filter and blood cell-bound activities, expressed as percentages of their respective maximum values, declined with t]/2 values of 4.7 and 4.9 h, respectively. Plasma activity is expressed as a percentage of the maximum (initial) whole blood activity.

loo

lOOOo 800-

E

~

~. 10

",7"

e o_

~ F[]

filters

bound

1

o

1%

i

2'o

Time f r o m injection

LT_

0

[] filters ------o---b,ood (~e.-boood~ -

lO I

I

I

4

6

I

;o

o

2'0

I

Time f r o m injection (h)

100-

8

Time from injection (h) Fig. 3. Time course of radiolabelled granulocyte migration into an abrasion 3 h old at the time of cell injection. Same symbols and axes as in Figs. 1 and 2. This subject undertook a 15 rain period of vigorous exercise (see materials and methods section), indicated by the arrow. The pre-exercise tl/2 values were 2.2 and 1.7 h for filter and blood cell-bound radioactivities respec -

tively,

'

200- ,

(h)

2

[]4h

400-

30

than the 8-h-old abrasion, the time courses of both were virtually identical with peak radioactivities at about 2 h after injection of the labelled cells (Fig. 4). The 4-h-old abrasion, however, showed a delayed peak, some 4 h after cell injection. We quantified the uptake of radioactivity into the filters as the ratio of filter radioactivity to the simultaneous cell-associated radioactivity in 1 ml whole blood. This ratio increased to a brief plateau value for each abrasion in each subject, m o r e

0

20 h

.

.c_ ~

Fig. 2. Time course of radiolabelled granulocyte migration into an abrasion 3 h old at the time of cell injection. Same symbols and axes as in Fig. 1. Filter radioactivity declined with a tl/2 of 7.4 h and blood cell-bound activity, 4.0 h.

-

~"~

600-

,

blOOd ( c e l l -

a.~

100.

b

U

b 0_

'

~

10.

,

filters (Bh)

1

0

5

10

15

2'0

2'5

Time f r o m injection ( h )

Fig. 4. a: time courses of radiolabelled granulocyte migration into three abrasions of different ages at the time of cell injection: (m) 20 h; ( , ) 8 h; (m) 4 h. Ordinate: counts per minute per filter per hour. The subject undertook a 15 rain period of vigorous exercise, indicated by the arrow, b: Time courses of radioactivity in the filters from the 8-h-old abrasion (O) and peripheral blood cells (D), expressed as percentages of the respective m a x i m u m values. Radioactivity declined with tl/2 values of 4.3 for filters and 4.0 h for cells.

rapidly in each abrasion of 8 h and older (Fig. 5). The maximum value for this ratio ranged from 0.3 in a 3-h-old abrasion to 1.32 for one of 20 h age at the time of cell injection. In the subjects illustrated in Figs. 3 and 4, exercise 4 - 6 h after cell injection produced a deflection in filter paper localisation curves. Thus, exercise caused a marked drop in filter radioactivity, which, after the termination of exercise, r e bounded before returning to its original rate of decrease. In all subjects plasma tll In activity remained essentially constant and very low, relative to the initial cell-associated radioactivity (Fig. 1).

10 Discussion This study shows that radiolabelled granulocytes are able to localise in abrasions of the type described by Addison et al. (1982). The technique of radiolabelling blood cells with n l I n is now well established. The label is very stable as indicated by the persistently low level of circulating cell-free nlln. Because of the relatively short residence time of granulocytes in the circulation, we expressed the plasma r a i n level as a percentage of the initial (i.e., maximal) whole blood r a i n activity rather than as a percentage of the simultaneous whole blood r a i n level. Otherwise, any changes in the actual plasma n l l n activity would be obscured, The possibility that the radioactivity in the filters simply represents exudation of r a i n labelled plasma proteins can be discounted by the finding of a low, constant plasma r a i n level and a complete discordance in the time courses of filter activity and corresponding plasma r a i n levels, Similarly, the possibility that the radioactivity in the filters represents actual bleeding can also be discounted on the basis of discordant time courses for blood and abrasions made less than 8 h before radiolabelled granulocyte injection (Figs. 2 and 3). Furthermore, the ratio of filter to blood radioactivity was far in excess of the volume of blood that the filter could absorb. Radioactivity in filters must, therefore, represent active radiolabelled granulocyte migration,

14-

"o 0

o



o

~

The localisation of granulocytes in the abrasion (' tissue' localisation) can be quantified as the ratio of filter to blood cell-bound radioactivity (Fig. 5). The filters were generally changed hourly and therefore this ratio has been expressed as the accumulation of radioactivity in the filters per hour. This ratio increased in 'mature' (see below) abrasions over the first few hours after radiolabelled cell injection, to reach a plateau which appeared to last a further few hours. It then tended to fall, presumably as a result of the increasing contribution to total cell-bound radioactivity of r a i n present in cells other than granulocytes, including cells which do not migrate. Thus, although about 95% of the injected r a i n was present on granulocytes, these cells have a short intravascular mean residence time of about 10 h and any non-granulocyte, cell-associated r a i n would represent an increasingly high fraction of total cell-bound n l l n . The plateau value for the filter to blood activity ratio would be the most appropriate quantitative index of tissue granulocyte localisation. The time course of filter radioactivity was essentially parallel to that of blood cell-bound radioactivity for abrasions made 8 h or more before labelled cell injection (Figs. 1 and 4), but was clearly displaced to the right of the blood r a i n curve for abrasions made earlier (e.g., Figs. 2 and 3), suggesting that tissue factors responsible for eliciting maximal tissue localisation of granulo-

20h

0.8-

b

0,6

1.o 0.8-



1

0 0

._3.

4h

....

0.6-

~

0.4

0.4-

u_

0.2

0.20.0

o

1'o

&

Time f r o m injection (I~)

i

0.0

o

1'o

2'0

Time f r o m injection (1~)

Fig. 5. Ratio of radioactivity in filter to that in 1 ml peripheral blood (cell-bound) as a function of time. a: filters from three abrasions in one subject: (13)20 h old at time of cell injection; (~) 8 h; (11)4 h. b: filters from abrasions in three other subjects: (t~), 20 h old at the time of cell injection; (~) 4 h; and (11)3 h.

11 cytes do not reach a steady state (i.e., that the abrasions do not 'mature') until 4 - 8 h after the abrasion. It has been previously demonstrated that exercise or adrenaline infusion causes granulocyte demargination in the microvasculature (Athens et al., 1961). The sensitivity of this technique was illustrated by the response of 'tissue' granulocyte localisation to exercise, which resulted in a perturbation of the time course of filter radioactivity in the subjects shown in Figs. 3 and 4. This suggested that exercise caused demarginatiOn of granulocytes at the sites of abrasions, with a resuiting decrease in granulocyte migration into the filters. It is conceivable that increased blood flow to muscle, with diversion from skin, was responsible for the observed perturbations, although this seems unlikely unless there was an actual decrease in skin blood flow. In conclusion, this novel approach which combines two techniques should allow the non-invasive quantification of tissue granulocyte localisation in man. Because the filters can be changed at will the approach permits repeated sampling over extended period of time. Thus, we can observe the effects of acute interventions (e.g., exercise) on ongoing localisation. Quantification of granulocyte migration in experimental animals, in contrast, has usually involved procedures, e.g. tissue dissection and radioactivity counting of samples

of inflamed tissue, which, by their very nature, terminate the experiment.

Acknowledgements A.M.P. was supported by the Cancer Research Campaign and A.J.P. and S.H.S. by the Wellcome Trust. We are grateful to Dr. I.E. Addison and the late Professor J.H. H u m p h r e y for help and advice. We are also grateful to our volunteers, without whose motivation these studies would not have been possible.

References Addison, I.E., Johnson, B. and Shaw, M. (1982) A human skin window technique using micropore membranes. J. Immunol. Methods 54, 129. Athens,J.W., Mauer, A.M., Ashenbrucker, H., Cartwright, G.E. and Wintrobe, M.M. (1961) Leukokinetic studies. III. The distribution of granulocytes in the blood of normal subjects. J. Clin. Invest. 39, 1481. Peters, A.M., Saverymuttu, S.H., Reavy, H.J., Danpure, H.J., Osman, S. and Lavender, J.P. (1983) Imaging inflammation with Ulln-tropolonate labelled leucocytes. J. Nucl. Med. 24, 39. Saverymuttu,S.H., Peters, A.M., Danpure, H.J., Reavy, H.J., Osman, S. and Lavender, J.P. (1983) Lung transit of Ill-indium labelled granulocytes. Relationship to labelling techniques. Scand. J. Haematol. 30, 151.

Quantification of radiolabelled granulocyte migration in vivo.

Quantification of the tissue localisation of granulocytes is difficult, particularly in man. With the aim of facilitating such quantification, a techn...
327KB Sizes 0 Downloads 0 Views