HistochemicalJournal, 7 (1975), 231-248

Microspectrophotometric detection of heparin in mast cells and basophilic granulocytes stained metachromatically with Toluidine Blue 0 .]'OHAN TAS

and

LIESBETH H. M. GEENEN

Histological Laboratory, University of Amsterdam, Iste Const. Huygensstraat 2% Amsterdam, The Netherlands

Received I November 1974

Synopsis. A qualitative microspectrophotometric detection method for heparin in situ

has been developed, using data obtained previously with a model system of polyacrylamide films containing pure glycosaminoglycans (Tas, 1975). This technique, based on the unique metachromatic properties of heparin with Toluidine Blue O in glycerol, has been worked out with rat peritoneal and mesenteric mast cells. After the smears containing the stained cells had been mounted in glycerol, a change with time of the recorded metachromatic peaks to lower wavelengths was found, leading to an equilibrium phase after some days. The metachromatic peaks recorded in this phase appeared to resemble closely the peak obtained for the heparin-Tohiidine Blue O complex under similar conditions in the model experiments. With rat mast cells it was found that nucleic acids, basic proteins, histamine and lipids had no appreciable influence on the position of the final recorded peaks, nor did they influence the slope of the time course very much. This observed decrease with time in the wavelengths of the metachromatic peaks can be explained by the time necessary for equilibration of the cells in glycerol and by the possible influence of lower sulphated glycosaminoglycans on the peak of the heparinToluidine Blue O complex. It was found that the method can be used to detect unequivocally the presence ofheparin in cells, even if they also contain up to 75% (mole/mole) of other, lower sulphated glycosaminoglycan. Only a limited number of cells is necessary with this m e t h o d - in contrast to biochemical determinations. For the first time the presence of heparin in normal human basophilic granulocytes and mast cells has been proved directly. The experiments indicate the occurrence of virtually similar sulphated heparins in human mast cells and basophilic granulocytes, as well as in pig mast cells. A higher sulphated heparin, however, might be present in rat mast cells.

Introduction

In 1878, Ehrlich reported that some kind of granular cell universally present in connective tissue stained specifically, and with very intense metachromasy, in the presence of 9 1975 Chapman andHallLtd 23I

23~

Tas and Geenen

a basic triaminotriphenylmethane dye called 'Dahlia'. Ehrlich considered that the granules of these cells consisted of accumulated reserve material. This is why he called the ceils 'Mastzellen' ('masten' in German = to fatten), a name which has been transferred into the English language as 'mast cells'~ About sixty years later, Jorpes and his co-workers contended for the first time that mast cells contain a snlphated, anticoagulating compound that stained metachromatically with Toluidine Blue, and which was called heparin (Holmgren & Wilander, I937; Jorpes et al., 1937). From this time, many investigators have analysed the composition of the mast cell and especially its granules, using a variety of biochemical and analytical techniques (for reviews, see Smith, 1963 and Selye, 1965). Mast cells from the rat contain not only heparin (3O-lOO ~g/lo 6 cells), but also histamine (lO-3o ~g/lo 6 cells), serotonin (5-HT; o.5-1 ~zg/Io6 cells) and protein. In addition, dopamine has sometimes been found, but heparin is virtually the only glycosaminoglycan (GAG) in normal adult mast cells (Selye, 1965; Hahn yon Dorsche et al., 197o). However, the presence of hyaluronic acid and chondroitin sulphate in mast cells has been reported (see Smith, 1963), but this has been found mostly under pathological conditions, e.g. in animal mastocytomas and human mastocytosis (e.g. Ringertz, 1963; Asboe-Hansen & Clausen, 1964; Lewis et al., 1973) or in cases of immature animal mast cells (e.g. Schiller, 1963). From electron microscopic studies, it is clear that the granules of e.g. rat peritoneal mast cells, which occupy over 5o% of the cytoplasmic volume (Helander & Bloom, I974), are surrounded by a distinct membrane and have initially a granular structure that becomes more electron dense and homogeneous during the maturation of the cells (Combs, I966). Different results have been reported for the molecular configuration of the content of mast cell granules. Using biochemical techniques, both Lagunoff and Uvn~is, with their respective colleagues, have gathered a great deal of evidence for the existence of an ionic complex between heparin-protein-histamine and 5-HT in the granules of the peritoneal mast cells of the rat. According to them, the first two compounds are bound by ionic interaction between the sulphate groups of the heparin and the amino groups of the protein. The latter are also bound electrostatically between the carboxyl groups of the protein and the amino group of histamine and 5-HT (Lagunoff et al., 1964; Aborg et al., 1967; Uvn/is et al., 197o; Bergqvist et al., 1971 ; Bergendorf& Uvniis, 1972). Evidence for the existence of a direct bond between heparin and the amines is the formation of granules in solutions containing heparin and different amounts of histamine, serotonin and basic protein (Csaba, 1971; Csaba & Ol~ih, 197~). The protein: heparin ratios (w/w) that have been reported so far vary between I : I (Lagunoff et al., 1964) and 1:2. 5 (Uvnas et al., 197o). The protein fraction of the rat peritoneal mast cell granules has been analysed chemically by Bergqvist et al. (1971) : it appeared to be a chymotrypsin-like protein with a molecular weight of 56o0 and an isoelectric point of about 9. Csaba & Surj~in (I97O), however, report that the basic protein in cultured rat thymus mast cells must be histone. Their assertion is based on ammonical silver staining for histones according to Black et al. (196o) and Black & Ansley (1964). However, these authors did not prove unequivocally that no other basic proteins can be stained by this procedure. As mentioned by Bergqvist et al. (197I), 'a minor part of the granule protein is cova-

Microspectrophotometric determination of heparin

233

lently bound to heparin', an assertion particularly defended by Lindahl (I966) and by Serafini-Fracassini et al. (1969, 197o). However, the mast cells investigated were isolated from pig intestinal mucosa and ox liver capsule, which can perhaps indicate the existence of differences between the physicochemical structure of mast cell granules from different sources As well as this major protein component with proteolytic properties, small amounts of mitochondrial and other enzymes have been detected in mast cell granules using organelle fractionatic procedures (Smith, I963; Selye, 1965). It is probable that this partly represents contamination of the granule fraction with mitochondrial and other components. Apart from chemical techniques, histochemical methods have also been applied to investigations of the nature of the glycosaminoglycans in developing mast cell granules. Not only has the metachromatic reaction with e.g. Toluidine Blue or Methylene Blue been widely used - in several cases at different pH levels and in the presence of cations (Landsmeer, I951; Schubert, I956; Moore & Schoenberg, 1957; Radden, I961, 196z; Parwaresh & Lennert, 1969; Chiu & Lagunoff, 197z; Leahu, x97z) - but also the Alcian Blue-Safranin O staining method according to Spicer (196o, 1963), e.g. by Combs et al. (1965) and Chiu & Lagunoff(i971, i972). Depending on the fixative used (Spicer, I963; Horsfield, I966), this combination of dyes stains sulphated glycosaminoglycans, other than heparin, blue, and heparin itself orange-red. Recently, Csaba (1971) has shown with model granules of known composition that the protein content, and particularly the amount of amines, also have an important influence on the results of this combined staining. Instead of Alcian Blue, Astra Blue can also be applied to the same end (Burton, 1964). Alcian Blue staining in the presence of MgC12 (Scott & Dorling, 1965) and a modified PAS technique (Scott & Harbinson, 1969; Scott & Dorling, 1969) have been applied by Butler (I97I). The presence of heparin as virtually the only glycosaminoglycan in the granules of normal adult mast cells is the conclusion of most of these histochemical investigations. Thus they support entirely the chemical findings which, however, have yielded more exact evidence in favour of this statement. Much less is known about the granules of the basophilic granulocytes in human or animal peripheral blood, long known to have much in common with mast cells. As described in reviews by Selye (1965) and Ackerman (1968), these granules also contain heparin, histamine, 5-HT and some protein. With regard to the glycosaminoglycans, Ackerman (1968), as well as nearly all the investigators mentioned by him, have used histochemical methods (e.g. Toluidine Blue metachromasy), histo-enzymatic techniques (e.g. hyaluronidase digestion) or other tests (e.g. blood clotting time) to 'demonstrate' the presence of heparin in the granules of human and animal basophilic granulocytes. The presence ofhyaluronic acid and chondroitin sulphate has been reported in the blood of some leukaemic patients. Only Amann & Martin (I96I) have carried out any chemical analyses. Using chromatographic methods, they analysed the glycosaminoglycan fraction isolated from basophilic granulocytes of patients with chronic myeloid leukaemia; it was concluded that heparin and perhaps some hyaluronic acid was present. As far as we have been able to find, no other analytical studies of this kind on human or animal basophilic granulocytes have been reported. The aim of this investigation was to determine exactly whether heparin is present in

234

Tas and Geenen

the granules of mast cells (rat, human) and basophilic leucocytes (human) by means of the metachromatic reaction with Toluidine Blue O. The knowledge that has been gathered in previous work with model experiments (Tas & Roozemond, I973; Tas, I975) has been used to develop a microspectrophotometric method for detecting heparin in metachromatically stained cells. Materials and m e t h o d s

Model film experiments Polyacrylamide films containing glycosaminoglycan (GAG), GAG and protein or a combination of two different GAGs were prepared as described by Tas & Roozemond (I973) and Tas (I975). The samples of purified GAGs and proteins used were the same as those described by Tas (I975). The staining and washing of the films, as well as the spectrophotometric recording procedures, have also been described previously (Tas & Roozemond, I973; Tas, I975). The films were stained with Toluidine Blue O bought from Serva, Heidelberg, West Germany (cat. no. 36692; dye content 93%) and dissolved in 20 mM citrate-phosphate buffer, pH 5.0 or pH 3.0 (o.oi-o.o3%, w/v).

Isolation of mast cells and basophilic granulocytes Peritoneal mast cells were washed out from the abdominal cavity of male and female white Wistar rats, as outlined by Thon & Uvn~is (I967). The animals, weighing 2oo25o g, were killed by decapitation; 4-5 ml isotonic Ringers' solution, adjusted to pH 6.o-6.2 with Io% o.i M citrate-phosphate buffer, pH 6.0, were then injected into the abdominal cavity. After careful massage for I min, the abdomen was opened, the intestines pushed aside and the peritoneal fluid was sucked off with a syringe, The collected fluid was centrifuged at i o o g for 2 min and decanted, and the white pellet was resuspended in about o.2 ml buffered Ringers' solution. Of this final suspension, cell smears were made and air-dried before fixation. Mast cells from rats were also investigated in the mesentery of the small intestine. As described by James (I96o), parts of the mesentery were stretched between two closely ftting, concentric, teflon rings before fixation. Using the same technique, parts of the mesentery were removed from the small intestine of three pigs. This was carried out at the slaughter house about ao min after the animals had been killed. As a source of human mast cells the mesentery of the vermiform appendix, the mesoappendix, was chosen. This material was obtained from six patients (two girls of 7 and I6 years, two women of 23 and 29 years and two men of I7 and 27 years) during an apFendectomy 'h froid'. To avoid the possibility that pathologically changed mast cells might be studied, material removed in cases of acute appendicitis was not used. Immediately after removing the appendix, it was dropped with the attached mesentery into the fixative solution. Then small portions of the delicate membrane were dissected free of fatty tissue and stretched on small cork plates for further processing. Human basophilic granulocytes were isolated from the venous blood of nine healthy volunteers (five women and four men, all aged 20 to 3~ years). The isolation technique was worked out by Drs H. Loos and A Ketel at the Central Laboratory of the Nether-

Microspectrophotometric determination of heparin

235

lands Red Cross Blood Transfusion Service, Amsterdam, The Netherlands (personal communication). They slightly modified the techniques described by B6yum (I968) and Day (I972). Blood samples (20 ml) were defibrinated by careful stirring with some glass beads (diameter o.5 cm). One part of the defibrinated blood (I 5 ml) was mixed with three parts of the tris-A solution (see Day, I972). This mixture was carefully layered over a Ficoll-Isopaque mixture (2 • I~ ml) with a density of I.O82 g/ml and an osmolarity of 290 mosmol, and centrifuged at a speed of4oo g for 3~ min at 2o~ The layer at the interphase region, containing only the mononuclear cells and the basophilic granulocytes, was carefully removed with some of the upper plasma layer but, as far as possible, without interfering with the Ficoll-Isopaque mixture. The fraction collected (about 8 ml) was washed three times by suspension in 4~ ml tris-A soIution and centrifugation at Ioo g for Io min at 2o~ The final pellet was suspended in about 0.2 ml tris-A solution and cell smears were prepared and air-dried before fixation. The final suspension appeared to contain only 2-5% basophilic granulocytes, instead of the i5-I8% reported by Day (z972). However, without modification, we obtained with his method a still smaller enrichment of basophilic granulocytes. Further purification by means of gradient centrifugation as outlined by Day (I972) was not carried out for these experiments, as our purpose was only to obtain an enrichment of some ten times for histochemical work.

Fixation, staining and enzymatic digestion of the cells The various cell smears and mast cell-containing membranes were fixed in methanol4 o ~ formaldehyde-acetic acid (85 :IO :5 by vol) for o.5/I hr. Rat peritoneal mast cells and rat mesenteries were also fixed in methanol (2 hr), absolute ethanol (z hr), Carnoy's fluid (o.5-2 hr), tetrahydrofuran-acetone (I :I by vol) for 0. 5 hr, 4o% formaldehyde vapour (o.5-2 hr), 90% ethanol, saturated picric acid-4o % formaldehyde-acetic acid (80 : 15:5 by vol) for I hr and distilled water, saturated picric acid-4o~ formaldehydeacetic acid (80:20:5 by vol) for 2 hr. After rinsing, the cells were stained with o.oi % Toluidine Blue O (Serva) dissolved in 20 mlvi citrate-phosphate buffer, pH 5.0, for 2- 5 min. In most cases, lipids were removed from the mesenteries before staining by immersing them in chloroformmethanol (2: I v/v) for 3~ min. Staining was followed by three 30 sec-I min rinses with distilled water. The cell smears were then air-dried, and mounted in glycerol. The stretched mesenteries were air-dried after being mounted on a slide. After the teflon rings had been carefully removed, the mesenteries were mounted in glycerol. The stretched meso-appendix mesenteries were also put on slides before air-drying and mounting in glycerol. In some experiments with rat peritoneal and mesenteric mast cells, several enzymatic digestions were carried out before staining. These cell smears and mesenteries were incubated at 37~ in one of the following enzyme solutions: (a) 0.5 mg deoxyribonuclease (Sigma, cat. no. DN-25 or Worthington, cat. no. D M L ; about Iooo i.u./mg) per ml o.o3 M MgSO4 in o.i M acetate buffer, pH 5.0 (b) I.o mg ribonuclease (Sigma cat. no. R 4875; about 75 i.u./mg) per ml o.I M acetate buffer, pH 5.0 (c) 2.0 mg pepsin (Sigma, cat. no. P 6875; about 25oo i.u./mg) per ml o.o2 N hydrochloric acid, pH 1.6

236

Tas and Geenen

(d) o.o2 ml papain suspension (Sigma, cat. no. P 3IZ5; about 48o i.u./ml) per ml o.2 M acetate buffer, pH 5.6, containing 4 mM EDTA and 2o mM L-cysteine-HC1 (Calbiochem, Lucerne, Switzerland) (e) o.I mg trypsin (Sigma, cat. no. T 8oo3; about 98oo i.u./mg) per ml o.o5 M phosphate buffer, pH 8.o or 2.5% (w/v) trypsin solution (Flow Laboratories, Maryland, USA, cat. no. 7-o2oB; activity unknown) diluted I : Io (v/v) with o.o5 M phosphate buffer, pH 8.o (f) 2.o mg histaminase (diamine oxidase; Sigma, cat. no. D 7876; about o.o6 i.u./mg) per ml o.I M phosphate buffer, pH 7.4 (g) I.o mg testicular hyaluronidase (Calbiochem, cat. no. 38593; about 5o0 i.u./mg) per ml o.i M phosphate buffer, pH 5.5 Control incubations in solutions without the enzymes were carried out simultaneously. The incubation time for (a), (b), (c), (f) and (g) was I hr, and for (d) and (e) I5-45 min The applied enzymes were washed out with distilled water before the slides were stained.

Microspectrophotometric recording After the cells had been stained with Toluidine Blue O and mounted in water-free glycerol (n~ ~ 1.455), spectra of the cell granules were recorded with a Zeiss universal microspectrophotometer UMSP I, using a pair of matching Ioo • Ultrafluar glycerol immersion objectives. The measuring spot used had a diameter of I g.m and the illuminating light spot one ofI 3 ~m. From the same specimens, spectra were recorded of the metachromatically stained cells 4-6 hr after mounting in glycerol and on subsequent days until peak changes were no longer observed and/or the stained cells had faded completely. The glycerol used as mounting medium was replaced every 2- 3 days, to avoid serious interference of orthochromatic staining with the metachromatic peaks. The slides were stored in the dark at 22~ With every measurement, spectra were recorded of the granules in five to ten randomly chosen cells of specimens from at least three different rats, pigs or humans. The mean of these peak wavelengths as well as the standard deviation was determined. Results As shown with a model system of polyacrylamide films by Tas & Roozemond (I973), it is precisely the peak of the heparin-Toluidine Blue O complex which takes in a particular position among the metachromatic complexes of this dye with different purified glycosaminoglycans (GAGs). The metachromatic shift becomes still more pronounced when this complex is equilibrated in a medium such as glycerol (Tas, I975)- It is because of these data that Toluidine Blue O and glycerol have been chosen for staining and mourning the mast cells and basophilic granulocytes used in this study. Preliminary experiments with rat peritoneal and mesenteric mast cells revealed that the metachromatic peaks recorded shortly after mounting the cells in glycerol, were found at wavelengths which were 20-30 nm (s.D. about 5 rim) above those of the peaks recorded with the model system (Tas, I975)o Because it is Well known now that the granules of these cells contain almost exclusively heparin (see Introduction for refer-

Microspectrophotometricdeterminationof heparin

z37

ences), some reasons must exist for this phenomenon of high 'first-day' peaks. Their position seemed to be independent of the fixative used in most cases Only after fixation with solutions containing formaldehyde and picric acid was a somewhat more elevated peak wavelength (by 5-IO nm) noticed, a finding in agreement with model experiments reported by Tas (I975). Therefore, for histochemical and morphological reasons methanol-4o % formaldehyde-acetic acid (85 :Io :5 by vol) was chosen as the routine fixative solution. For the enzymatic digestions, however, the cells were mostly fixed with absolute ethanol or in Carnoy's fluid (Pearse, I972). The digestions with deoxyribonuclease, pepsin, histaminase or testicular hyaluronidase did not significantly change the 'first-day' peaks. On the contrary, ribonuclease lowered these peaks somewhat (by 5-Io nm). This was also the case with trypsin, even after fixation with 4o% formaldehyde vapour. However, incubation with trypsin for I hr or longer completely destroyed the cells. With papain, the cells were destroyed even faster. The control incubations (without the enzymes) never showed any influence on the 'first-day' peaks. With model experiments too, slight influences of electrostatically bound proteins on the metachromatic peaks in an aqueous medium as well as in glycerol have been found (Tas, I975)- Additional experiments have shown that after incubation of the GAGprotein films in several fixatives~ 4% formaldehyde, absolute or 6o)/o methanol-4o % formaldehyde-acetic acid (85 :IO :5 by vol) and Carnoy's fluid respectively for I hr, the metachromatic peaks with Toluidine Blue O were not further influenced. In mast cells in the mesentery, lipids also did not seem to be an important factor in determining the peaks; we found that a 3o min rise in chloroform-methanol (z :I by vol) did not alter the peaks significantly. The collagenous fibres in the mesenteries showed no absorption either. The experiments mentioned above did not completely clarify the observed phenomenon of the high 'first-day' peaks. Apart from some influences of RNA and proteins on the metachromatic peaks, there may be other explanations for this phenomenon. This was confirmed by the finding that the wavelengths of the recorded metachromatic peaks from granules continuously decreased during the first few days after they had been mounted in glycerol. As shown in Fig. x for rat peritoneal mast cells fixed in absolute

I

WAVELENGTH OF METACHROMAIICPEAK in nm

Figure I. Time course of metachromatic peak changes of rat peritoneal mast cells~ stained with Toluidine Blue O and mounted in glycerol. T h e cells were fixed in absolute ethanol for ~ hr. Staining time 5 min. T h e lower curve ( 9 represents cells incubated with o.z5 ~o trypsin solution (3o rain) before staining. T h e upper curve ( e ~ O ) represents untreated cells. Temperature of storing the cells zz~

53O

520

510 nrnI

510

1

511nm

-

-

-

-

4

i

Z 3 4 5 6 TIME AFTER MOUNTINGIN GLYCEROL(DAYS)

238

Tas and Geenen

ethanol, the wavelength of the recorded peaks reaches an equilibrium after some days, with a minimal value of about 511 nm. As already mentioned above, trypsin digestion lowers the peaks recorded shortly after the mounting in glycerol, but it does not alter the final result. However, after trypsin digestion, the granules were stained less intensely and faded at greater speed than the untreated ones~ Other enzyme digestions (deoxyribonuclease, ribonuclease, histaminase, testicular hyaluronidase) did not influence the final result of Fig. I either. Metachromatically stained rat peritoneal mast cells shortly after mounting in glycerol and cells in the equilibrium phase (where they are faded) are illustrated in the photomicrographs of Fig. 2. Two metachromatic curves ( - . - . - ) having somewhat different

Figure 2. Photomicrographs of Toluidine Blue O stained rat peritoneal mast cells as used for

Fig. I. (a) Cells I day after mounting in glycerol; (b) the same cells six days later, photographed under similar conditions. • II5o peak wavelengths, which were recorded from rat mast cells, are shown in Fig. 3- For comparison, the curves of a nucleus, of a model film and of Toluidine Blue O dissolved in glycerol, are also plotted in this figure. Results similar to Fig. I were obtained with rat mesenteric mast cells, likewise fixed 1.3t EXTINCTION

/'--~

117 .9 .8 .7 .6

//

//

/"

,/.:

.5 .4 .3

/// ,,/ ..............

.2 .1

..........-Y i.60 480 500 520 540 560 580 600 620 6/.*0 660 WAVELENGTHin nm

Figure 3. Metachromatic and orthochromatic spectra of Toluidine Blue O (TBO) recorded in vitro and in situ.

.......

TBO dissolved in glycerol, ~ heparin film stained with TBO and equilibrated in glycerol, ..... two microspectrophotometric curves from rat peritoneal mast cells, stained with TBO and mounted in glycerol, ...... microspectrophotometric curve from a rat lymphocyte nucleus, stained with TBO and mounted in glycerol. -

-

Microspectrophotometric determination of heparin

239

in absolute ethanol (Fig. 4). In contrast to the free peritoneal mast cells, the mast cells in the mesenterium needed a period which was 3-5 times longer before reaching the equilibrium level. The application of some other fixatives on peritoneal and mesenteric mast cells delivered peak courses quite similar to Figs. i and 4 respectively, with final peak wavelengths of again about 51 o n m (see Table I). All these final values are in rather good agreement with the position of the peak as obtained for the heparin-Toluidine Blue O complex in glycerol in the model experiments. The small differences between these two findings will be discussed later. WAVELENGTH OF METACHROMATIC PEAK in nrn

540

530

520

510

T,ME A~TE~ .00N,,NG ,N

Figure 4. Time course of metachromatic peak changes of rat mesenteric mast cells, stained with Toluidine Blue O and mounted in glycerol. The cells were fixed in absolute ethanol for 2 hr. Staining time z rain. The lower curve (C)--C)) represents cells incubated with o.25% trypsin solution (3o rain) before staining. The upper curve (0---0) represents untreated cells. Temperature of storing the cells 22~ In view of the original problem concerning the high 'first-day' peaks, the changes in the metachromatic peaks with time remain to be clarified. On the one hand, the stained, air-dried cells must equilibrate in the non-aqueous mounting medium. With stained, air-dried heparin films we found that the equilibrium of these films in glycerol required several days (depending on temperature) and was coupled with a drop in the wavelength of the metachromatic peak. A similar time-requiring process can be expected to occur in situ too. In contrast, we obtained evidence that when rat mast cell granules still contain some other GAG, apart from heparin, this might also influence the time course of the peaks. As can be illustrated with the model experiments of Table 2, this will lead to metachromatic peaks at higher wavelengths in glycerol, but not before 30% (by vol; molar ratio heparin-chondroitin sulphate is then about 5 :I) or more sulphated G A G is present. Otherwise, the disturbance of the peak of the heparin-Toluidine Blue O complex by e.g.

Tas and Geenen

240 ~8

,-,

,-.,

I

o

o

o

,-,

/ +

[

,-,

[ +

,-,

o,.-, ~

] +

,-,

,-,

I

I

I

oo

oo

o~

i

~

U

~A

~9

~A

(.9

0

0,.~ "~

0

22

4-

"~

~

"~

~

~

o~

Microspectrophotometric determination of heparin

24I

Table 2. Metachromatic peaks of polyacrylamide films containing heparin (Hep) and different amounts of chondroitin 4-sulphate (C 4-S). The films have been stained with o.oi % Toluidine Blue O, dissolved in 2o mM citrate-phosphate buffer, pH 5.o, for 2o min. After rinsing in distilled water (three times 3 min), a part of the films has been equilibrated in glycerol as described by Tas (1975). The peaks showed to have a variation of I-3 nm.

Percentage of C 4-S to ttep (w/w) o

Molar ratio Hep/C 4-S

Metachromatie peak as recorded in distilled water (nm)

Metachromatic peak as recorded in glycerol(nm)

516 516 516 517 52i 531

--

527

5 io 20 30 4o

42:I is:i 8:i 5:1 3:1

527 528 532 534 536

50

2:1 1:1

7~ 85

1:3

538

535

539 54o

538 543

heparan sulphate or chondroitin sulphate is dependent on time. Fig. 5 illustrates that the dissociation of the h e p a r i n - T o l u i d i n e Blue O complex in glycerol declines considerably more slowly than complexes containing lower sulphated G A G s . T h e dissociation of the complex with dextran sulphate, being about 1.5 times more sulphated than heparin, declines the slowest (metachromatic peaks in glycerol and distilled water at wavelengths o f 5 IO and 5o7 n m respectively). I n distilled water and 2 mM citrate-phosphate buffer, p H 5.o, the fastest dissociation has been found for hyaluronic acid followed by chondroitin 4- and 6-sulphate, keratan sulphate, heparan sulphate and heparin, respectively. I n other words, with the passage o f time after mounting, the influence of (small) amounts of lower sulphated G A G s on the peak of the h e p a r i n - T o l u i d i n e Blue O complex will gradually be cancelled out before all metachromasy is faded. T h e results of experiments with polyacrylamide films containing heparin plus different amounts of chondroitin sulphate (Fig. 6) support this view. Even in the presence of 86% (w/w) chonEXTINGTION AT ME TACHRONATIC PEA~ 1,2.

1.0 S

0.8.

I

0,6.

04.

0.2C4-S and HS 0.0

2~

ie

~

9~ TIME IN HOURS

Figure 5- Time course of dissociation of different polyacrylamide films, stained with Toluidine Blue O. The experiments were performed as described by Tas & Roozemond (1973) in I. 5 ml glycerol. Temperature xo~ The different enclosed acid polysaccharides (5 mg per ml polymerization solution) are indicated with the following abbreviations: C 4-S -- chondroitin 4-sulphate; HS = heparan sulphate; HEP = heparin; DEX S = dextran sulphate.

242

Tas and Geenen WAVELENGTH OF METACHROMATIC PEAK in rtm

Figure 6. Time course of metachromatic peak changes of polyacrylamide films containing heparin and different amounts of chondroitin 4-sulphate, stained with Toluidine Blue O and equilibrated in glycerol. The films were stored in 1.5 ml glycerol at 2o~ The glycerol medium was stirred before every reading and was refreshed every 24 hr. The dots plotted in the graph are the mean of three experiments. The molar ratios of the enclosed heparin and chondroitin &sulphate mixtures were: 3: I ( I - - i ) ,

5AO

530

520

510 I

r

t

i

24

i

48

2:I (O--O),

i

72 TIME IN HOURS

I:I (O--O),

and I: 3

( D - - ~ ) respectively.

droitin sulphate (molar ratio heparin-chondroitin sulphate is then about I : 3) the peak of the heparin-Toluidine Blue O complex is evident before all metachromasy has disappeared. T h e actual importance of this concept for the experiments described in this paper will be discussed later on. The recording technique developed during the experiments with the rat mast cells was applied to cells from other sources. Because the commercial heparin that is used in the model experiments is isolated from pig intestines, the mesenteric mast cells of this animal were investigated. A very good correlation between the findings obtained with microspectrophotometry and those found with the model system is evident (see Tab]e i). Fig. 7 shows the time course of the recording peaks of metachromatically stained human mast cells. For comparison, the results concerning rat mesenteric mast cells that have been fixed similarly, have been plotted in the same figure. Both types of mast cells show a long period (about i o days) before reaching the equilibrium level. Finally, the results obtained with normal human basophilic granulocytes are presented l WAVELENGTH OF HETACHROMATIC PEAK in nrn 56O 550 540

530 520 513ninth__ 509nm --

510

1

2

3

4

5

6

7

8

9

10

11

~2

13

14

15

16 17 18 19 20 TI~.~E AFTER MOUNTING IN GLYCEROL (DAYS)

Figure 7. Time course of metachromatic peak changes of human mast cells from meso-appendix mesenteries ( O - - O ) and rat mesenteric mast cells ( i - - m ) , stained with Toluidine Blue O and mounted in glycerol. The cells were fixed in methanol-4o % formaldehyde-acetic acid (85:zo:5 by vol) for 89hr. Staining time 5 min. Temperature of storing the cells 2z~

Microspectrophotometric determination of heparhz OF 530 WAVELENGTH METACHROMATIC PEAKin nm

~2o I t

516nm

I

510 1

2

3

-4

5

5

7

243

Figure8. Time course of metachromatic peak changes of human basophilic granulocytes, stained with Toluidine Blue O and mounted in glycerol. The cells were fixed in methanol-4o% formaldehyde-acetic acid (85:1o :5 by vol) for 89 hr. Staining time 5 rain. Temperature of storing the cells 22~

8

TIMEAFTERMOUNTINGIN GLYCEROL(DAYS)

in Fig. 8. With these cells, special care was taken to avoid interference of the orthochromatically stained nucleus with the metachromatic peaks as much as possible. This had to be done because in these cells the nucleus mostly occupies the main part of the total cell volume (Fig. 9a), except in rare cases of cell damage on the slides (Fig. 9b).

Figure9. Photomicrographs of Toluidine Blue O stained human basophilic granulocytes as used for Fig. 8. (a) Undamaged ce!lin the smear; (b) A cell, probably damaged by the smearing procedure. x 2240

As can be seen from Table I, the final metachromatic peaks of both human cell types are much alike. In contrast to the situation with the animal cells investigated, different preparations of human cells showed a much wider variation in dissociation time of the complexes with Toluidine Blue O. The final peak wavelengths of the individual preparations, however, all ended at the same final wavelength.

Discussion

On the basis of the model experiments of Tas & Roozemond (I973), it can be stated that the appearance of metachromasy in mast cells and basophiIic granulocytes forms no proof for the presence ofheparin exclusively in these cells. On the other hand, the appli-

244

Tas and Geenen

cation ofmicrospectrophotometry to these metachromatically stained cells has made the development of a specific detection method for heparin possible. This method is based on the exceptional position of the metachromatic peak of the heparin-Toluidine Blue O complex in glycerol (Tas, 1975). In general, it represents a histochemical application of the knowledge gathered previously with model experiments, that also has made understandable many of the histochemical problems encountered in the present investigation. Firstly, the choice of the routine fixation solution is based in part on the results of the model experiments in this respect (Tas, I975). More precisely than with stained cells, the model experiments have shown that fixation of heparin and heparan sulphate with solutions containing formaldehyde and picric acid has to be avoided because some kind of degradation of the GAGs occur under these conditions. Furthermore, depending on the protein content of the granules, care has to be taken with all aqueous fixation solutions (Parwaresh & Lennert, I967; Selye, I965). These 'warnings', combined with additional morphological findings, have made it possible to select a fixative suitable for routine purposes: the methanol-formaldehyde-acetic acid mixture. Although this fixative solution is stated by Pearse (I972) to be unsuitable for enzymatic digestions, our experiments on the incubation of rat mast cells with some protein hydrolysing enzymes appeared to be an exception to this statement: fixation with formaldehyde vapour instead of the advised absolute ethanol or Carnoy's fluid (Pearse, I972) did not block these enzymatic digestions very much. With polyacrylamide films containing GAG plus protein, it has been shown that electrostatically bound proteins only slightly influence the metachromatic peaks with Toluidine Blue O (Tas, I975). This may clarify the observed lowering of the 'first-day' peaks by 5-IO nm after trypsin digestion of the rat mast cells. The same model experiments have shown that under the conditions as used in this investigation, the presence of chymotrypsin-like proteins (selected according to Bergqvist et al., I97I) does not diminish the degree of dye binding by heparin. On the basis of the cytological relationships within the cell, the interference of orthochromatically staining nuclear DNA and RNA with the metachromatic peaks from the mast cell granules is not very likely. The RNA content of the mast cell cytoplasm, shown by u.v. microspectrophotometry to be very low (Tas, unpublished observation), also cannot be an important influencing factor. In our opinion, the reason for the observed small peak lowering caused by the digestion with this enzyme will be the protease activity of the not highly pure ribonuclease sample that we used. The absence of an effect of histaminase digestion on the position of the 'first-day' peaks seems to be in favour of the granule configuration model of Uvn~s and his group (see Introduction) which includes no ionic binding between heparin and histamine. However, the potency of histamine to compete with Toluidine Blue O might be very small, so that the analogy with the model of Uvn~is and his co-workers may not be justified. The absence of an effect of testicular hyaluronidase digestion on the 'first-day' peaks seems to support the statement that normal rat mast cells contain virtually only heparin and no other glycosaminoglycan(see Introduction for references). Furthermore, Table 2 shows that the metachromatic peak of heparin in glycerol is not significantly changed until 3o% (w/w) or more of lower sulphated GAG is present besides heparin. With regard to the possible influence of (phospho) lipids, it should be pointed out that their presence inside the granules of normal rat mast cells is still questionable (see Selye,

Microspectrophotometric determination of heparin

245

I965). Extraction of the granule and/or extracellular lipids has not indicated any kind of hindrance of the complex formation of heparin with Toluidine Blue O by these compounds, if they are present anyhow. The facts gathered so far leave only a few explanations possible for the phenomenon of the time course of appearance of the metachromatic peaks, as has been recorded from the different cells (Figs. I, 4, 7, 8). On the one hand, it is beyond doubt, in our opinion, that the time necessary for the equilibration of the stained cells in the mounting medium (glycerol) is a very important cause of the drop in the wavelength of the metachromatic peaks during the first days. The compact network of intercellular substance as present in the mesenterium and meso-appendix will cause a slower penetration of glycerol into the mast cells in these membranes than in the case of smears of free lying cells. More complex, on the other hand, is the influence of lower sulphated GAGs on the time course. On the basis of various findings, namely (a) Figs. 5 and 6 of this work, (b) the protein blocking experiments of Tas (I975) showing that lower sulphated GAGs are more readily blocked for dye binding than heparin, and (c) the experiments of Toepfer (I973, I974), which reveal similar molar extinction coefficients for the different sulphated GAGs, it can be concluded, on the whole, that the influence of lower sulphated GAGs on the metachromatic peak of the heparin-Toluidine Blue O complex in glycerol will be cancelled out, as long as the amount of these lower sulphated GAGs does not exceed the amount of heparin by more than about three times (mole/mole). In these cases, the peak wavelengths in the equilibrium phase will be similar to that of the heparin-Toluidine Blue O complex (about 517 nm). When a still smaller part of the granule GAG is heparin, the peak wavelengths at the equilibrium phase will correspond closely to the peak of the complexes of Toluidine Blue O with e.g. chondroitin sulphate or heparan sulphate (about 545 nm). Besides the time of equilibration mentioned above, the slopes of the time courses, or in other words the time necessary to reach the equilibrium phase, as well as the time that this equilibrium persists (until the cells are faded away) will depend on the amounts of the different GAGs, temperature and also on the influence of light (Kelly & Bloom, I959, and our own observations). Both temperature and light cause a more rapid dissociation ofmetachromatic complexes. Looking again at Figs. I, 4, 7 and 8, it can be stated now that the presence of heparin in the granules of the different cells investigated, has been proved with this technique. From the results, summarized in Table I, a striking similarity between the model experiments and the cytological objects appears. Only the rat mast ceils reveal a significantly lower mean peak wavelength than is obtained with the heparin-Toluidine Blue O complex in glycerol with the model experiments. On the basis of the present results, it cannot be considered very likely that lower sulphated GAGs play a role in this phenomenon. It seems more probable that an explanation should be looked for in the presence of a somewhat higher sulphated heparin in the mast cells of the rat, as compared with the heparin in the mast cells of pig and human. Support for this view can be found in the lower metachromatic peak with Toluidine Blue O in glycerol for dextran sulphate than for heparin and moreover in the results of chemical analyses of rat skin and pig mucosal heparin by Homer (I97o). In the data obtained by Homer, a higher total sulphate: hexosamine molar ratio was found for the former than for the latter (about 2.9 and 2.6 respectively). It is because of the fact that various techniques are available for assaying sulphate and hexosamine, that simultaneous and similar determinations of their molar

246

Tas and Geenen

ratios of different types of heparin must be done to make the data comparable. Therefore, the varying molar ratios of total sulphate to hexosamine, as determined by different investigators (Parekh & Glick, 1961; Schiller, 1963; Kavanagh & Jaques, 1973; and Mathews in Tas, 1975) cannot be used. In conclusion, this investigation has shown that the microspectrophotometrical determination of heparin described gives a specific proof of the presence of this compound even if the investigated object contains over 75% (mole/mole) of other lower sulphated GAGs apart from heparin. The technique can be applied when only a limited number of animal or human cells is available, present in e.g. smears or tissue sections or tissue whole mount preparations. Quantitative data about the amounts of heparin or lower sulphated GAGs cannot be obtained with this method. Only relative amounts might be estimated roughly in principle (from the slopes of the time courses). Kelly & Bloom have clearly shown already in 1959, that the quantitative microspectrophotometric determination of Toluidine Blue metachromasy is virtually impossible. On the other hand, these authors have shown that taking the sum of the ortho- and metachromatic peak extinctions as a measure of the amount of bound dye is more promising. The technique described in this paper might yield some information about the degree of sulphation of the heparin concerned. This possibility is at present being investigated.

Acknowledgements The authors wish to thank Professor J. James, Dr R. C. Roozemond and Professor J. M. Tager for their encouraging discussions in performing the experiments and for their help concerning the preparation of this paper. We are very grateful to Professor W. H. Brummelkamp (Hospital Binnengasthuis, Amsterdam) and Dr J. J. M. Schroeder (Lucas Hospital, Amsterdam) and their colleagues for the willing assistance in obtaining the different samples of the mesentery of the vermiform appendix.

References ABORG, C. H., NOVOTNY, J. • UVNAS, B. (1967). Ionic binding of histamine in mast cell granules. Acta physiol, scand. 69, 276-83. ACKERMAN, G. A. (1968). Cytochemical properties of the blood basophilic granulocyte. Ann. N. Y. Acad. Sci. lO3, 376-93 . AMANN, R. & MARTIN, H. (I96I). Blutmastzellen und Heparin. Acta haemat. 25, 2o9-19. ASBOE-I-IANSEN, G. & CLAUSEN, J. (1964). Mastocytosis (Urticaria pigmentosa) with urinary excretion ofhyaluronic acid and chondroitin sulfuric acid. Am.J. Med. 36, 144-5o. BERGENDORFF, A. & UVNAS, B. (1972). Storage of 5-hydroxytryptamine in rat mast cells. Evidence for an ionic binding to carboxyl groups in a granule heparin-protein complex. Acta physiol, scan& 84, 320-31 9 BERGQVIST, U., SAMUELSSON, G. & UVNAS, B. (197t). Chemical composition of basophil granules from isolated rat mast cells. Actaphysiol. scand. 83, 362-72. BLACK, M. M. & ANSLEu H. R. (1964). Histone staining with ammoniacal silver. Science x43, 693-5. BLACK~M. M., SPEER~ F. D. & LILLICK~ L. C. (1960). Acid-extractable nuclear proteins of cancer cells. I. Staining with ammouiacal silver.J, ham. Gancer Inst. 25, 967-77. B6YUM, A. (1968). Isolation of mononuclear cells and granulocytes from human blood. Scan& J. Clin. Lab. Invest. 2I, Suppl. 97, 77-89 9

Microspectrophotometric determination of heparin

247

BURT ON, A. L. (1964). H i s t o c h e m i c a l studies o n developing m a s t cells. Anat. Rec. 15o , 2 6 5 - 9. BUTLER, W. F. (I97I). Variations in t h e staining o f c u t a n e o u s m a s t cells. Histoehem. ft. 3, 365 7~ 9 CHIU, H. & LAGUNOFP, D. (I97I), H i s t o c h e m i c a l c o m p a r i s o n o f frog a n d rat m a s t cells.J. Histochem. Cytochem. I9, 369-75 ` CHIU, H. & LAGUNOFP, D. (I972). H i s t o c h e m i c a l c o m p a r i s o n o f vertebrate m a s t cells. Histochem. .7. 4, 135-44. c o MB S, 3. W. (I966). M a t u r a t i o n of rat m a s t ceils. A n electron microscopic study. J. Cell Biol. 3 I, 563-75 . c o M B S, J. w., L A6 • N O r r, n . & B E N D I T T, F.. p. (1965.) Differentiation a n d proliferation o f e m b r y onic m a s t cells o f t h e r a t . J . Cell Biol. 25~ 577-92. CSABA, G. (1971). M e c h a n i s m o f the f o r m a t i o n o f m a s t cell granules. V I I . Participation o f a m i n e s a n d basic proteins in t h e f o r m a t i o n o f t h e m a s t cell granule. Analysis o f the heterogeneity of m a s t cells. Acta biol. Acad. Sci. hung. zz, I55-68. CSAI3A, G. & OLAH, I. (1972). M e c h a n i s m o f t h e f o r m a t i o n o f m a s t ceil granules. V I I I . U l t r a s t r u c t u r e o f m o d e l granules. Acta biol. Acad. Sci. hung. 23, I33-44. CSABA~ G. & 5URJAN, L. (I970). M e c h a n i s m o f t h e f o r m a t i o n o f m a s t cell granules. V. H i s t o chemical detection o f histones in t h e m a s t cell granules. Actu biol. Acad. Sci. hung. zI, 3 4 I - 6 . DAY, R. P. (1972). Basophil leucocyte separation f r o m h u m a n peripheral blood: a t e c h n i q u e for their isolation in h i g h p u r i t y a n d h i g h yield. Clin. Allergy, z, 2o5-12. :EH R L I C H ~ V. ( 1878 ). B eitrage z u r T h e o r i e u n d Praxis dee histologischen F~irbung. Thesis. U n i v e r sity o f Leipzig. HAHN VON DORSCHE, t{., EEHRMANN, P. • SULZMANN, R. (1970). D i e Mastzelle als einzellige e n d o k r i n e Drfise. Actu anat. 77, 56o-9. HELANDER, H. F. & BLOOM, G. D. (I974.) Quantitative analysis o f m a s t cell s t r u c t u r e . f t . Microscopy Ioo, 315--2I. HOL2dGREN, H. & WILANDER, O. (I937). Beitrag z u r K e n n t n i s dee C h e m i e u n d F u n k f i o n der E h r l i c h s c h c n Mastzellen. Z. miler, anat. Forsch. 42, 242-78. HORSPIELD, G. I. (1966). M a s t cell mucopolysaccharides. Nature 211, 422. 5AMES, 5. (I960). O b s e r v a t i o n s on t h e so-called sex c h r o m a t i n . Z. Zel!forsch. mikrosk. Anat. 51, 597-616. 5ORPES, E., HOLMGREN, H. & WILAND ER, O. (1937). U b e r das V o r k o m m e n y o n H e p a r i n in d e n Gefiil~w~nden u n d in d e n A u g e n . Z. miler, anat. l~brsch. 4 z, 279-3oo. KAVANAGH, L. W. & JAQUES, L. B. (1973). C o m p a r i s o n o f analytical values for c o m m e r c i a l heparin. Arzneim. Porsch. (Drug Res.) z3, 6 o 5 - I I. I