Cell Tiss. Res. 190, 459-479 (1978)
Cell and Tissue Research 9 by Springer-Verlag 1978
The Effect of Haemorrhage on the Cell Populations of the Thymus and Bone Marrow in Wild Starlings (Sturnus vulgaris) Marion D. Kendall* Department of Anatomy, St. Thomas's Hospital Medical School, London, England
Summary. Following the withdrawal of blood from the brachial vein of adult wild starlings (Sturnus vulgaris) changes in the cell populations within the bone m a r r o w and thymus were observed over an eight day period. The packed cell volume, haemoglobin content and reticulocyte count of the peripheral blood was determined before and after haemorrhage. The m a x i m u m effect of the haemorrhage was observed in the bone m a r r o w after four days when the population of small lymphocytes, and basophilic erythroid precursors were reduced to less than 1%. At the same time the percentage of another line of erythroid cells increased to 68 %. This second erythroid lineage was the major erythroid line in the thymus, and again m a x i m u m representation occurred at 4 days post haemorrhage. After this the thymus became predominantly lymphoid and started to increase in size. The two erythroid lines are described and their status with regard to avian thrombocytes is also discussed. The peripheral blood had not attained the pre-haemorrhagic values for reticulocyte counts by eight days although the packed cell volumes and haemoglobin contents were similar. Key words: Thymus - Bone marrow - Birds - Erythropoiesis - Haemorrhage.
Introduction Previous studies on the thymus glands of the red billed Quelea (Quelea quelea) have shown that at certain times of the year the thymic lobes become the site for the development of red blood cells (Kendall and Ward, 1974; W a r d and Kendall, 1975;
Send offprint requests to: Dr. M.D. Kendall, Department of Anatomy, St. Thomas's Hospital Medical School, Lambeth Palace Road, London, SE1 7EH, England * I would like to thank Dr. Peter Ward of the Institute of Terrestrial Ecologyfor help in obtaining the starlings. Thanks are also due to the staff of the Anatomy Department of St. Thomas's Hospital Medical School, and in particular Mr. Watson. This and other work on the thymus is possibledue to the support of the Research (Endowments) Committee of St. Thomas's Hospital
0302-766X/78/0190/0459/$04.20
460
M.D. Kendall
Bacchus a n d Kendall, 1975; Kendall, 1975; Kendall, 1975a). N o comparative studies were done then o n t h y m u s a n d b o n e m a r r o w , b u t it seemed possible from pilot studies that the t h y m u s would r e s p o n d to h a e m o r r h a g e by e n l a r g e m e n t a n d erythropoiesis. A l t h o u g h the n a t u r a l a n d induced responses m a y n o t be the same this study o n starlings was u n d e r t a k e n to investigate these two responses. The starlings were t a k e n from the wild at a time (overwintering) when the birds were neither breeding n o r moulting. T h u s it was h o p e d that, in most cases, the thymuses w o u l d be small a n d regressed.
Materials and Methods Adult overwinteringstarlings (Sturnus vulgaris)were obtained from the wild in February 1977 and kept in a large open air aviary for 4 weeks. They were then transferred to individual cages and allowed to acclimatise to the new surroundings for 1 week. After this period of time 0.25 to 0.3 ml of blood was withdrawn from the brachial vein of each bird. Estimates were made of the packed cell volume (microhaematocrit technique), haemoglobin content (cyanmethaemoglobinmethod) and reticulocytecount (percentageof cellswith new methyleneblue vital stain reaction). Smears made from cells incubated for 20 min with new methylene blue were, after performing the counts, stained with the Lephene's method for haemoglobin (Undritz, 1973). Some birds were killed immediately after the initial withdrawal of blood; further samples were taken from others at 2 day intervals for 8 days. Prior to death, blood was again withdrawn from the brachial vein and the estimates of packed cellvolume, haemoglobin content and reticulocytecount were repeated. Immediately after death some of the thymic lobes were excised and homogenisedwith autologous or homologous (pigeon) plasma. Smears were prepared from the resultant suspension and stained with one of the following;Lephene's stain for haemoglobin followedby Leishman's stain, PeriodicAcid Schiff (P.A.S.) and haematoxylin (McManus, 1946), Giemsa or Wright's blood stain. Differential counts were performed on 500 cellstaken in a line traversing the slide. These differentialcounts excludedcellsjudged as reticulocytes by their rounded and slightly leptochromatic nucleus, and mature erythrocytes. The numbers of these were however recorded at the same time. Other thymic lobes were excised and fixed in either Zenker or Bouin's fixatives. These specimens were embedded in paraffin, cut at 5 txm and stained with haematoxylin and eosin, Dominici's stain, Krag's Kiton red-almond green method for erythrocytes (Lendrum, 1949), Leishman's stain and the amido black technique for haemoglobin (Puchtler and Sweat, 1962). Smears and paraffin embedded stained sections were also prepared similarly from the bone marrow of the left femur.
Results A t the time of death the 16 birds that survived the h a e m o r r h a g e weighed 66.69+ 6.31 gms. a n d the 3 that died d u r i n g the experiment weighed 51.66+ 1.56 gins. ( P < 0.001).
Peripheral Blood The i n d i v i d u a l results for each o f the birds examined over the 8 day period are s h o w n in Fig. 1. They have n o t been aggregated to emphasise the degree of variation observed. Three birds died d u r i n g the experiment, a n d it can be seen that each of these birds h a d at least one variable that c o r r e s p o n d e d closely to those o f the experimental birds after haemorrhage.
Effect of Haemorrhage on Thymus and Bone Marrow of Starlings
FPew
~
461
El:irl
4~
:/lll 3 0 - RETICULOCYTES Olo 20 IO O
- .AI I da2s5 Day O DIED
Da~s 0 0 2
111'11 11 0o4
Oo5
Oo8
Fig. 1. The packed cell volume(PCV),haemoglobincontact (lib) and reticulocytecount for each of the birds examined. The open rectangles [] denotevalues obtained from blood taken on day 0, and the solid rectangles 9 are the correspondingpost-haemorrhagevalues. Note that for PCV and Hb the y axis shows the physiological range only as higher and lower values have not been observed in normal starlings O f the birds that survived the haemorrhage, all had initial packed cell volumes of 46-57 ~ . The values after the haemorrhage were at their lowest at two and four days, and approximately regained their initial volume by six to eight days posthaemorrhage. Birds with a large percentage loss in packed cell volume were not necessarily those with the greatest drop in haemoglobin content. As with packed cell volumes, initial values were approximately regained by 6 days post-haemorrhage. The initial reticulocyte counts were all below 4.4 % with the majority below 2.4 ~ ; the counts after haemorrhage were slightly raised on the second day, and were much higher for the rest of the experiment. Examination of the smears prepared after mixing the cells with new methylene blue showed that the high counts were associated with the release into the blood of cell forms less mature than those seen in the initial smears or those of the second day. All the reticulocytes and mature red blood cells gave a positive reaction for haemoglobin using Lephene's stain.
Bone Marrow Histology It was not possible to quantify the amount of bone marrow present in any of the birds, but observations were made on a plug of tissue removed from some
462
M.D. Kendall
specimens. These samples were sectioned, and stained with a variety of stains (see Materials and Methods). Each of the specimens examined (7) contained fat cells interspersed with sinusoids and foci of developing erythrocytes, granulocytes, lymphoid cells and plasma cells. Lymphoid cells in each specimen were aggregated into distinct lymphoid follicles. In the three specimens taken from starlings before bleeding the follicles were the least developed. The largest follicle observed measured 200 ~tm in diameter, but one specimen had none. The follicles in these birds appeared predominantly lympoid with the only 'haemoglobin-positive' cells being mature erythrocytes in small capillaries within the follicles. The degree of vascularisation of the follicles was slight, so that there were only a few erythrocytes visible. The other four specimens were taken from birds 6-8 days after bleeding. Here the size of the largest lymphoid follicles was approximately 270 ~tm in diameter and there were many more follicles than in the previous group. Moreover, many nuclei within the lymphoid follicles of the second group were slightly to moderately positive for the haemoglobin stains. In addition there were a number of granulocytes on the periphery of the follicles from the second group that had 'haemoglobin-positive' granules present in the cytoplasm. Examination of the non-lymphoid areas of the bone marrow of both groups showed that the haemoglobin stains differentiated the erythroid cells into those in which the reaction was first observed in the nucleus with only a pale reaction from the cytoplasm, and cells where the cytoplasmic positivity preceded the nuclear reaction. These two different lines of erythroid cells were better observed in the smears. All stages of the development of granulocytes were observed, and mature plasma cells were identified.
Thymus Sizes and Histology All of the lobes on both sides of the neck were examined and an overall assessment was made of the size and colour of the lobes (Table 1). Previous sampling from starlings in the area showed that the thymus lobes of most adult starlings were small or slightly enlarged during the overwintering period. Sometimes, however, some pink or red lobes were observed (Ward, personal communication). Histologically, all of the small lobes examined had a distinct cortex and medulla and contained small but distinct Hassall's corpuscles. They did not therefore resemble the regressed small lobes previously found in Quelea quelea (Bacchus and Kendall, 1975). With an increase in size there was an increase in the numbers of mitotic figures found in the cortex, and a clear shift in nuclear size and morphology from the small dense pachychromatic thymocyte to a larger more leptochromatic cell. All of the 3-4 mm long thymic lobes had large numbers of red blood cells both in the blood vessels and free in the glands. Both staining techniques for haemoglobin showed numerous cells in the day 6 and day 8 specimens that gave a strong positive reaction in the nucleus. Even from cells in the blood vessels, the
Effect of Haemorrhage on Thymus and Bone Marrow of Starlings
463
Table 1. To show the increase in size and change of colour that occurred in the thymus lobes after haemorrhage. Most birds had 14-16 thymic lobes disposed as two chains, one on either side of the neck Days
0
2
4
6
8
Bird No.
Length of lobes, and colour 1-2 mm
2-3 mm
1 2 3
All white All white Most white
1 pink
4 5 6
Only 1 white found All white Most white
1 white
Some white Some white
Most white and pink Most pink Some white
8 9 10 7 12 16
3-4 mm
Some white
17 19
4+ mm
3 pink All pink Some pink Most white Most white
1 red, 1 white 1 white, 2 pink
1 red, 1 pink 1 red
haemoglobin reaction of the cytoplasm was less strong than in the nucleus. In contrast a few circulating cells had a strong positive reaction in the cytoplasm and a negative nuclear reaction.
The Cells of the Bone Marrow and Thymus In order to assess the relative proportions of distinct cell types in the smears prepared from thymus and bone marrow suspensions, the cells were counted differentially and then categorised by the criteria outlined below. This approach had its limitations; apart from the difficulty of cell identification at the light microscope level there was rarely a sharp dividing line between the features of one cell category and the next in ceils belonging to a developmental sequence, thus the limits of the categories were difficult to define. Also, particularly following haemorrhage, there was an alteration in the rate of development of some features such as basophilia or haemoglobin. Cells with such anomalies were difficult to categorise but very useful for the observations concerning cell lineages.
Lymphoid Cells Pyknotic Cells. Small cells (approximate diameter 4 gm) with distinctly pachychromatic nuclei that normally stained deeply with basophilic dyes. Some cells contained unstained spheres within the nuclei. Cytoplasm scanty and difficult to observe.
464
M . D . Kendall
Fig. 2. The cells of the bone marrow before haemorrhage. 1 Medium lymphocyte; 2 Fairly mature basophilic erythrocyte (reticulocyte); 3 Acidophilic erythroid cell with clear cytoplasm; 4 Mature erythrocyte; 5 Acidophilic erythroid cell with some Lephene reaction; 6 and 7 Two cells that are difficult to place in differential counts. The basophilia of the cytoplasm and lack of Lephene reaction suggests a lymphoid cell, yet the cell shape, nucleus and similarities with cells 3 and 5 suggest early erythroid cells of the acidophilic lineage; 8 Reticulocyte; 9 Basophilic erythroblast with slightly green cytoplasm due to a combination of Lephene + ve material and basophilia; 10 Large lymphocyte; probably a lymphoblast. (Barline = 10 lam)
Fig. 3A-I. The cells of the bone marrow and thymus of starlings. All photographs magnified x 1,260, and most prepared from Leishman and Lephene stained smears. A and B. Bone marrow, edge of smear. Two days post-haemorrhage. 1 Very slightly immature erythrocyte; 2 Mature erythrocyte; 3 Plasma cell; 4 Transitional cell; 5 Macrophage; 6 Small lymphocyte; 7 Large basophilic erythroblast in mitosis; 8 Smaller basophilic erythroblast; 9 Damaged cells; 10 Macrophage containing ingested erythrocytes; 11 Medium lymphocyte; 12 Basophilic erythroblast; 13 Heterophil; 14 Small lymphocyte. C Thymus. Four days post-haemorrhage. 15 Transitional cell. I)--F. Bone Marrow. Before haemorrhage. 16 Acidophilic erythroid cell with peripheral granules. These peripheral granules were acidophilic and relatively uncommonly seen. Here however they serve to show the extent of the clear cytoplasm around the dense acidophilic nucleus. (Giemsa stain); 17 A group of basophilic erythroblasts; 18 Erythrocyte; 19 Basophilic erythroblasts with one in mitosis. G and H. Bone marrow. Six days post-haemorrhage. 20 Basophilic erythroblast; 21 Fairly late basophilic erythroblast with weak + ve reaction to Lephene's stain; 22 Acidophilic erythroid cell with some Lephene + ve reaction; 23 A similar cell to 22 with less Lephene + ve reaction; 24 Acidophilic erythroid cell with basophilia and slight Lephene + ve reaction in cytoplasm. I Thymus. Four days post-haemorrhage. 25 A very similar cell to 24
Effect of Haemorrhage on Thymus and Bone Marrow of Starlings
465
466
M.D. Kendall
Small Lymphocytes. (Fig. 3, 6 and 14; Fig. 6, 1). Cells less than 6 pm in total diameter. The cytoplasm was scanty and normally polar in position. The nuclei were distinctly pachychromatic with a nucleolus usually visible. The cytoplasm stained slightly with basophilic dyes. These include the thymocytes of the thymus.
Larger Lymphoid Cells. (Fig. 2, 1; Fig. 3, 4, 6, 11 and 15; Fig. 6, 2, 3 and 4). A heterogeneous group of cells all of which were greater than 6.5 ~tm in diameter and had cytoplasm clearly visible round the nucleus. The nucleus was typically rounded but the degree of leptochromasia within allowed some cells to be distinguished as transitional cells (Rosse, 1976), others as medium or large lymphocytes and yet other as possible blast cells. Transitional cells had a small amount of cytoplasm that was normally basophilic but the degree of basophilia varied in different cells. The nucleus was clearly leptochromatic in large cells and less so in smaller cells. One or two nucleoli were normally present. Medium and large lymphocytes were similar in size to the transitional cells but the nucleus was always to some extent pachychromatic, and nucleoli were readily discernable. The cytoplasm was normally weakly basophilic and usually more extensive than in transitional cells. Occasional strongly acidophilic granules were observed in the cytoplasm. Possible blast cells had more cytoplasm than either lymphocytes or transitional cells. The nucleus was rounded and generally smaller than medium or large lymphocytes (typically approximately 6 pm diameter). There was one clear, often centrally placed nucleolus and sometimes possibly one other. Cells containing more than two nucleoli were regarded as medium or large lymphocytes. The nucleolus was ring-shaped and comparatively large. Nuclear chromatin was often aggregated peripherally as shown in Fig. 6, 5, but in some instances this aggregation was incomplete so that the nucleus appeared leptochromatic in part. The more leptochromatic the nucleus the more acidophilic it appeared. The cytoplasm was normally visible all round the nucleus and was slightly basophilic. These possible blast cells formed an unbroken spectrum of cells from a small dense pachychromatic cell with little cytoplasm (a small lymphocyte) to the acidophilic erythroid line described later. To facilitate quantitative observations on cell populations, all cells in this series with basophilic cytoplasm were classified as lymphoid, and if there was no cytoplasmic basophilia visible then the cells were classified as erythroid. The complete spectrum of cells is shown in (Fig. 6). Cells resembling Fig. 6, 7 were occasionally observed as doublets immediately after mitosis.
Erythroid Cells The developmental sequence of erythroid cells in chicken bone marrow, according to Lucas and Jamroz (1961), involves a large rounded basophilic cell, the erythroblast, that is progressively reduced in size whilst still retaining the rounded leptochromatic nucleus and small amount of basophilic cytoplasm until there is a hint of haemoglobin formation in the cytoplasm. At this stage the cell (that may
Effect of Haemorrhage on Thymus and Bone Marrow of Starlings
467
measure about 10 ~tm) is termed a polychromatic erythrocyte. Their drawings show very little basophilia of the cytoplasm at this stage. When the full haemoglobin colour is visible in the cytoplasm the nucleus has often become more oval. At this stage too the typical oval cell outline is formed and further reduction in nuclear size and degree of leptochromasia leads to the formation of the mature form. The thrombocytic line of development in chickens is depicted as differing from the erythrocyte line in that the initial basophilia of the cytoplasm is greater and persists for longer, the nucleus has a different (punctate) chromatin pattern although it is still leptochromatic and no haemoglobin develops. The nucleus of the mature thrombocyte is larger than that of the erythrocyte, and is a violet purple colour rather than a bluish colour with 'Romanowsky-type' stains. Cytoplasmic granules persist and are regarded as characteristic of mature circulating thrombocytes. Detailed examination of the starling smears prepared in this study showed differences that were not reconcilable with the above classical picture for developing avian erythrocytes or thrombocytes. Firstly the basophilic erythroblast cells of Lucas and Jamroz (1961) were easily observed but the degree of cytoplasmic basophilia was very intense and persisted for a long time after haemoglobinisation was observed. (This may be of little consequence as it is known that the timing of the development of haemoglobin in the erythroid cells can vary according to different physiological conditions). Secondly haemoglobin was seen to develop in the cytoplasm of cells that contained no basophilia whatsoever, were already ovoid in shape, had larger acidophilic nuclei than the erythroid cells, and were in all respects, except for a lack of cytoplasmic inclusions, thrombocytes. Thirdly, these "thrombocytes" develop from precursor cells that were only very faintly basophilic in the largest most primitive stage. They rapidly lost any basophilia so that the cytoplasm was very difficult to observe, even under phase contrast, and had throughout a very acidophilic nucleus. Small azurophilic granules were sometimes observed with Giemsa staining. Acidophilia was retained even when the cytoplasm appeared fully haemoglobinised, but it must have been lost eventually as all erythrocytes looked the same in circulating peripheral blood. Thus in this study the erythroid cells have been regarded as either basophilic (with basophilic cytoplasm) or acidophilic (with acidophilic nuclei). Combining the 'Romanowsky-type' stains with Lephene's stain for haemoglobin allows the following categories to be identified.
Basophilic, Non-Haemoglobinised Cells. (Fig. 3, 7, 8 and 12). These ranged in size from very large cells (10 Ixm diameter) to cells approximately 6 ~tm in overall diameter. Both the cells and the nuclei were rounded. The nucleus was leptochromatic, occasionally with a nucleolus. The cytoplasm was distinctly basophilic. Mitotic figures were observed, particularly in the larger cells.
Basophilic, Haemoglobinised Cells. (Figs. 2, 9; Fig.'3, 17, 19, 20 and 21). In general, smaller in size than the non-haemoglobinised cells. Following haemorrhage some cells were very small indeed (4-5 ~tm). The cytoplasm ranged from slightly green in colour, with a lot of basophilia still present, though a distinctly green stage with a smaller nucleus to a more brownish colour of the cytoplasm as the haemoglobin
468
M.D. Kendall
0 ]'''''.
~
ii
li P
60
20
B1
0
Z
.............
1O~o 0
2
4
,~ A2
~
~EO
6
8
[
DAYS
Fig. 4. Results of the differential counts performed on bone marrow smears. The lymphoid compartment (top). L-L Larger lymphoid cells (medium and large lymphocytes, transitionals and possible blast cells excluding basophilic erythroblasts); S-L small lymphocytes; P pyknotic cells. The erythroid compartment (higher middle). A1 Acidophilic erythroid cells; A2 more mature acidophilic erythroid cells; B1 Basophilic erythroid cells; B2 more mature basophilic erythroid cells (often green in colour because of some degree of Lephene reaction). Damaged cells (D) and miscellaneous cells (M) miscellaneous cells include macrophages, promyelocytes and plasma ceils. The myeloid compartment (bottom); EO eosinophils of both sizes; N Heterophils. The basophils were less than 0.5 ~o predominated. The intensity o f the brown colour was frequently reduced following haemorrhage. Mitosis was c o m m o n l y observed. The nucleus was round, and often contained acidophilic granules in the nucleoplasm. Lephene positive material was observed between the heterochromatin, sometimes giving a stronger reaction than in the cytoplasm. W h e n basophilia o f the cytoplasm was largely lost, Lephene + ve material predominated then these cells were regarded as reticulocytes. Acidophilic, Non Haemoglobinised Cells(Fig. 2, 3; Fig. 3, 16 and 23; Fig. 6, 6 and 7). The distinguishing feature o f these cells was the acidophilic nucleus, that had very
Effect of Haemorrhage on Thymus and Bone Marrow of Starlings
469
S-L
40
Z2O
............
-
.
-
'~ L - L
~P
40
Z20 AI A2
Z 2~ ...........
0
i ....
2
. . . . . . . . . ~. . . . . . . . . . . .
4
n. . . . . . . . . . . .
6
=M
8
DAYS
Fig. 5. Results of the differential counts performed on thymus smears. Abbreviations as for Fig. 4. The percentages of myeloid cells was very low and similar in all counts
little differentiation in all but the larger, presumably more primitive, cells. A pale nucleolus was sometimes seen. The cytoplasm was usually unstained and very difficult to distinguish even under phase contrast microscopy as it rarely contained any inclusions. Mitotic figures were observed. Some of the cells appeared very fragile and it was not all uncommon to find cells in which the cytoplasm was partially unstained and partially acidophilic as though nucleoplasm had streamed out from the nucleus. Lucas and Jamroz (1961) show a figure of a similar but vacuolated cell that they regard as a technical artifact. This may be so, but it is a useful occurrence as it enables the oval cell outline to be observed at least in part of the cell. Occasionally the unstained part was either faintly basophilic or more commonly, slightly 'Lephene-positive'. These acidophilic cells have been designated A1 cells in Figs. 4 and 5. A more mature form, described below was designated A2.
470
M.D. Kendall
5
5 pro,
Fig. 6. A selection of drawings of cells from the thymus 4 days post-haemorrhage. 1 Small pachychromatic lymphocyte; 2 Medium lymphocyte with a less pachychromatic nucleus (early small transitional); 3 Transitional cell. Occasionally transitional cells had very basophilic cytoplasm and then they resembled basophilic erythroblasts; 4 Possible blast cell of the acidophilic erythroid lineage. The amount of cytoplasm is variable; 5 Similar to 4 sometimes but the slightly basophilic cytoplasm can become "clear" in places; 6 Acidophilic erythroid cell with clear cytoplasm and a slight Lephene + ve reaction. The inclusions at the top are only sometimes present and they range in number from 3-12; 7 Cells with acidophilic nuclei and clear cytoplasm that are either in mitosis or close to it. No cytoplasmic inclusions or basophilia are apparent
The r o u n d e d more primitive cells appeared to be linked morphologically with large l y m p h o i d cells clearly seen 4--6 days after h a e m o r r h a g e in both the thymus and in the bone marrow. (See l y m p h o i d cell descriptions.) Smaller, m o r e dense, but still acidophilic nuclei were associated with an oval cell shape, similar to that o f an erythrocyte (A2 cells). N o r m a l l y these cells lacked any inclusions but otherwise would have been regarded as thrombocytes. Frequently, however, there was a very slight positive reaction to Lephene's stain for haemoglobin.
Acidophilic, Haemoglobinised Cells. (Fig. 2, 5; Fig. 3, 22, 24 and 25). The acidophilic fairly h o m o g e n o u s , oval nucleus can be clearly seen within the 'Lephene-positive' cytoplasm. N o inclusions or basophilia were observed in the cytoplasm. These cells were included in the counts o f reticulocytes (with the basophilic lineage reticulocytes described above).
Thromboeytes. Previous authors (Lucas and Jamroz, 1961; M u k k u r and Bradley, 1967; Santamarina, 1964) have found that thrombocytes can be confused with either p o o r l y developed erythrocytes or lymphocytes. The use o f P.A.S. to stain glycogen in the cytoplasm (Mills Westerman, 1974) m a y help with the identification.
Effect of Haemorrhage on Thymus and Bone Marrow of Starlings
471
Table 2. To show the percentage of mature erythrocytes, reticulocytes and thrombocytes that contained cytoplasmic granules. Mature erythrocytes should be Lephene + ve and not contain granules whereas thrombocytes should be Lephene - v e and contain cytoplasmic granules Day
Lephene + ve No granules
Granules
Lephene + ve+ Basophilic (green)
Lephene - ve No granules
Granules
0
61.5_+ 7.1
16.5_+ 5.4
10.2+10.0
11.7_+10.2
0.2+0.3
2
78.3_+ 3.4
7.8-+ 5.6
7.2_+ 6.2
6,7_+ 1.0
4
34.2_+28.7
1.5_+ 2.6
0.8_+ 1.4
63,3_+30.6
6
14.8_+ 7.1
13.8_+ 11.7
64.3_+ 6.4
7.0__+ 10.8
0
8
76.7-+ 14.5
4.3_+ 3.9
14.2_+20.1
4.0+ 2.8
0.8-+ 1.1
0
66.5-+ 7.5
22.3-+ 4.9
0
10.5-+ 9.8
0.7-+0.8
2
86.5-+11.6
5.3_+ 4.3
0
7.8+ 7.6
0.2+0.3
4
44.7_+ 9.5
2.3+ 0.8
0
52.3+10.0
0.7-+0.6
Bone marrow
0 0.2+0.3
Thymus
*
~
***
6
73.3_+ 9.5
17.0-+ 9.5
0
9.7+ 1.5
0
8
83.5_+ 4.9
12.5_+ 7.8
0
3.5_+ 3.5
0.5_+0.7
93.8_+ 4.7
6.1+ 4.7
0
0
0
90.6_+ 15.9
8.5_+ 13.0
0
0
0
Peripheral blood 0 All posthaemorrhage. *
P=
Cell and Tissue Research 9 by Springer-Verlag 1978
The Effect of Haemorrhage on the Cell Populations of the Thymus and Bone Marrow in Wild Starlings (Sturnus vulgaris) Marion D. Kendall* Department of Anatomy, St. Thomas's Hospital Medical School, London, England
Summary. Following the withdrawal of blood from the brachial vein of adult wild starlings (Sturnus vulgaris) changes in the cell populations within the bone m a r r o w and thymus were observed over an eight day period. The packed cell volume, haemoglobin content and reticulocyte count of the peripheral blood was determined before and after haemorrhage. The m a x i m u m effect of the haemorrhage was observed in the bone m a r r o w after four days when the population of small lymphocytes, and basophilic erythroid precursors were reduced to less than 1%. At the same time the percentage of another line of erythroid cells increased to 68 %. This second erythroid lineage was the major erythroid line in the thymus, and again m a x i m u m representation occurred at 4 days post haemorrhage. After this the thymus became predominantly lymphoid and started to increase in size. The two erythroid lines are described and their status with regard to avian thrombocytes is also discussed. The peripheral blood had not attained the pre-haemorrhagic values for reticulocyte counts by eight days although the packed cell volumes and haemoglobin contents were similar. Key words: Thymus - Bone marrow - Birds - Erythropoiesis - Haemorrhage.
Introduction Previous studies on the thymus glands of the red billed Quelea (Quelea quelea) have shown that at certain times of the year the thymic lobes become the site for the development of red blood cells (Kendall and Ward, 1974; W a r d and Kendall, 1975;
Send offprint requests to: Dr. M.D. Kendall, Department of Anatomy, St. Thomas's Hospital Medical School, Lambeth Palace Road, London, SE1 7EH, England * I would like to thank Dr. Peter Ward of the Institute of Terrestrial Ecologyfor help in obtaining the starlings. Thanks are also due to the staff of the Anatomy Department of St. Thomas's Hospital Medical School, and in particular Mr. Watson. This and other work on the thymus is possibledue to the support of the Research (Endowments) Committee of St. Thomas's Hospital
0302-766X/78/0190/0459/$04.20
460
M.D. Kendall
Bacchus a n d Kendall, 1975; Kendall, 1975; Kendall, 1975a). N o comparative studies were done then o n t h y m u s a n d b o n e m a r r o w , b u t it seemed possible from pilot studies that the t h y m u s would r e s p o n d to h a e m o r r h a g e by e n l a r g e m e n t a n d erythropoiesis. A l t h o u g h the n a t u r a l a n d induced responses m a y n o t be the same this study o n starlings was u n d e r t a k e n to investigate these two responses. The starlings were t a k e n from the wild at a time (overwintering) when the birds were neither breeding n o r moulting. T h u s it was h o p e d that, in most cases, the thymuses w o u l d be small a n d regressed.
Materials and Methods Adult overwinteringstarlings (Sturnus vulgaris)were obtained from the wild in February 1977 and kept in a large open air aviary for 4 weeks. They were then transferred to individual cages and allowed to acclimatise to the new surroundings for 1 week. After this period of time 0.25 to 0.3 ml of blood was withdrawn from the brachial vein of each bird. Estimates were made of the packed cell volume (microhaematocrit technique), haemoglobin content (cyanmethaemoglobinmethod) and reticulocytecount (percentageof cellswith new methyleneblue vital stain reaction). Smears made from cells incubated for 20 min with new methylene blue were, after performing the counts, stained with the Lephene's method for haemoglobin (Undritz, 1973). Some birds were killed immediately after the initial withdrawal of blood; further samples were taken from others at 2 day intervals for 8 days. Prior to death, blood was again withdrawn from the brachial vein and the estimates of packed cellvolume, haemoglobin content and reticulocytecount were repeated. Immediately after death some of the thymic lobes were excised and homogenisedwith autologous or homologous (pigeon) plasma. Smears were prepared from the resultant suspension and stained with one of the following;Lephene's stain for haemoglobin followedby Leishman's stain, PeriodicAcid Schiff (P.A.S.) and haematoxylin (McManus, 1946), Giemsa or Wright's blood stain. Differential counts were performed on 500 cellstaken in a line traversing the slide. These differentialcounts excludedcellsjudged as reticulocytes by their rounded and slightly leptochromatic nucleus, and mature erythrocytes. The numbers of these were however recorded at the same time. Other thymic lobes were excised and fixed in either Zenker or Bouin's fixatives. These specimens were embedded in paraffin, cut at 5 txm and stained with haematoxylin and eosin, Dominici's stain, Krag's Kiton red-almond green method for erythrocytes (Lendrum, 1949), Leishman's stain and the amido black technique for haemoglobin (Puchtler and Sweat, 1962). Smears and paraffin embedded stained sections were also prepared similarly from the bone marrow of the left femur.
Results A t the time of death the 16 birds that survived the h a e m o r r h a g e weighed 66.69+ 6.31 gms. a n d the 3 that died d u r i n g the experiment weighed 51.66+ 1.56 gins. ( P < 0.001).
Peripheral Blood The i n d i v i d u a l results for each o f the birds examined over the 8 day period are s h o w n in Fig. 1. They have n o t been aggregated to emphasise the degree of variation observed. Three birds died d u r i n g the experiment, a n d it can be seen that each of these birds h a d at least one variable that c o r r e s p o n d e d closely to those o f the experimental birds after haemorrhage.
Effect of Haemorrhage on Thymus and Bone Marrow of Starlings
FPew
~
461
El:irl
4~
:/lll 3 0 - RETICULOCYTES Olo 20 IO O
- .AI I da2s5 Day O DIED
Da~s 0 0 2
111'11 11 0o4
Oo5
Oo8
Fig. 1. The packed cell volume(PCV),haemoglobincontact (lib) and reticulocytecount for each of the birds examined. The open rectangles [] denotevalues obtained from blood taken on day 0, and the solid rectangles 9 are the correspondingpost-haemorrhagevalues. Note that for PCV and Hb the y axis shows the physiological range only as higher and lower values have not been observed in normal starlings O f the birds that survived the haemorrhage, all had initial packed cell volumes of 46-57 ~ . The values after the haemorrhage were at their lowest at two and four days, and approximately regained their initial volume by six to eight days posthaemorrhage. Birds with a large percentage loss in packed cell volume were not necessarily those with the greatest drop in haemoglobin content. As with packed cell volumes, initial values were approximately regained by 6 days post-haemorrhage. The initial reticulocyte counts were all below 4.4 % with the majority below 2.4 ~ ; the counts after haemorrhage were slightly raised on the second day, and were much higher for the rest of the experiment. Examination of the smears prepared after mixing the cells with new methylene blue showed that the high counts were associated with the release into the blood of cell forms less mature than those seen in the initial smears or those of the second day. All the reticulocytes and mature red blood cells gave a positive reaction for haemoglobin using Lephene's stain.
Bone Marrow Histology It was not possible to quantify the amount of bone marrow present in any of the birds, but observations were made on a plug of tissue removed from some
462
M.D. Kendall
specimens. These samples were sectioned, and stained with a variety of stains (see Materials and Methods). Each of the specimens examined (7) contained fat cells interspersed with sinusoids and foci of developing erythrocytes, granulocytes, lymphoid cells and plasma cells. Lymphoid cells in each specimen were aggregated into distinct lymphoid follicles. In the three specimens taken from starlings before bleeding the follicles were the least developed. The largest follicle observed measured 200 ~tm in diameter, but one specimen had none. The follicles in these birds appeared predominantly lympoid with the only 'haemoglobin-positive' cells being mature erythrocytes in small capillaries within the follicles. The degree of vascularisation of the follicles was slight, so that there were only a few erythrocytes visible. The other four specimens were taken from birds 6-8 days after bleeding. Here the size of the largest lymphoid follicles was approximately 270 ~tm in diameter and there were many more follicles than in the previous group. Moreover, many nuclei within the lymphoid follicles of the second group were slightly to moderately positive for the haemoglobin stains. In addition there were a number of granulocytes on the periphery of the follicles from the second group that had 'haemoglobin-positive' granules present in the cytoplasm. Examination of the non-lymphoid areas of the bone marrow of both groups showed that the haemoglobin stains differentiated the erythroid cells into those in which the reaction was first observed in the nucleus with only a pale reaction from the cytoplasm, and cells where the cytoplasmic positivity preceded the nuclear reaction. These two different lines of erythroid cells were better observed in the smears. All stages of the development of granulocytes were observed, and mature plasma cells were identified.
Thymus Sizes and Histology All of the lobes on both sides of the neck were examined and an overall assessment was made of the size and colour of the lobes (Table 1). Previous sampling from starlings in the area showed that the thymus lobes of most adult starlings were small or slightly enlarged during the overwintering period. Sometimes, however, some pink or red lobes were observed (Ward, personal communication). Histologically, all of the small lobes examined had a distinct cortex and medulla and contained small but distinct Hassall's corpuscles. They did not therefore resemble the regressed small lobes previously found in Quelea quelea (Bacchus and Kendall, 1975). With an increase in size there was an increase in the numbers of mitotic figures found in the cortex, and a clear shift in nuclear size and morphology from the small dense pachychromatic thymocyte to a larger more leptochromatic cell. All of the 3-4 mm long thymic lobes had large numbers of red blood cells both in the blood vessels and free in the glands. Both staining techniques for haemoglobin showed numerous cells in the day 6 and day 8 specimens that gave a strong positive reaction in the nucleus. Even from cells in the blood vessels, the
Effect of Haemorrhage on Thymus and Bone Marrow of Starlings
463
Table 1. To show the increase in size and change of colour that occurred in the thymus lobes after haemorrhage. Most birds had 14-16 thymic lobes disposed as two chains, one on either side of the neck Days
0
2
4
6
8
Bird No.
Length of lobes, and colour 1-2 mm
2-3 mm
1 2 3
All white All white Most white
1 pink
4 5 6
Only 1 white found All white Most white
1 white
Some white Some white
Most white and pink Most pink Some white
8 9 10 7 12 16
3-4 mm
Some white
17 19
4+ mm
3 pink All pink Some pink Most white Most white
1 red, 1 white 1 white, 2 pink
1 red, 1 pink 1 red
haemoglobin reaction of the cytoplasm was less strong than in the nucleus. In contrast a few circulating cells had a strong positive reaction in the cytoplasm and a negative nuclear reaction.
The Cells of the Bone Marrow and Thymus In order to assess the relative proportions of distinct cell types in the smears prepared from thymus and bone marrow suspensions, the cells were counted differentially and then categorised by the criteria outlined below. This approach had its limitations; apart from the difficulty of cell identification at the light microscope level there was rarely a sharp dividing line between the features of one cell category and the next in ceils belonging to a developmental sequence, thus the limits of the categories were difficult to define. Also, particularly following haemorrhage, there was an alteration in the rate of development of some features such as basophilia or haemoglobin. Cells with such anomalies were difficult to categorise but very useful for the observations concerning cell lineages.
Lymphoid Cells Pyknotic Cells. Small cells (approximate diameter 4 gm) with distinctly pachychromatic nuclei that normally stained deeply with basophilic dyes. Some cells contained unstained spheres within the nuclei. Cytoplasm scanty and difficult to observe.
464
M . D . Kendall
Fig. 2. The cells of the bone marrow before haemorrhage. 1 Medium lymphocyte; 2 Fairly mature basophilic erythrocyte (reticulocyte); 3 Acidophilic erythroid cell with clear cytoplasm; 4 Mature erythrocyte; 5 Acidophilic erythroid cell with some Lephene reaction; 6 and 7 Two cells that are difficult to place in differential counts. The basophilia of the cytoplasm and lack of Lephene reaction suggests a lymphoid cell, yet the cell shape, nucleus and similarities with cells 3 and 5 suggest early erythroid cells of the acidophilic lineage; 8 Reticulocyte; 9 Basophilic erythroblast with slightly green cytoplasm due to a combination of Lephene + ve material and basophilia; 10 Large lymphocyte; probably a lymphoblast. (Barline = 10 lam)
Fig. 3A-I. The cells of the bone marrow and thymus of starlings. All photographs magnified x 1,260, and most prepared from Leishman and Lephene stained smears. A and B. Bone marrow, edge of smear. Two days post-haemorrhage. 1 Very slightly immature erythrocyte; 2 Mature erythrocyte; 3 Plasma cell; 4 Transitional cell; 5 Macrophage; 6 Small lymphocyte; 7 Large basophilic erythroblast in mitosis; 8 Smaller basophilic erythroblast; 9 Damaged cells; 10 Macrophage containing ingested erythrocytes; 11 Medium lymphocyte; 12 Basophilic erythroblast; 13 Heterophil; 14 Small lymphocyte. C Thymus. Four days post-haemorrhage. 15 Transitional cell. I)--F. Bone Marrow. Before haemorrhage. 16 Acidophilic erythroid cell with peripheral granules. These peripheral granules were acidophilic and relatively uncommonly seen. Here however they serve to show the extent of the clear cytoplasm around the dense acidophilic nucleus. (Giemsa stain); 17 A group of basophilic erythroblasts; 18 Erythrocyte; 19 Basophilic erythroblasts with one in mitosis. G and H. Bone marrow. Six days post-haemorrhage. 20 Basophilic erythroblast; 21 Fairly late basophilic erythroblast with weak + ve reaction to Lephene's stain; 22 Acidophilic erythroid cell with some Lephene + ve reaction; 23 A similar cell to 22 with less Lephene + ve reaction; 24 Acidophilic erythroid cell with basophilia and slight Lephene + ve reaction in cytoplasm. I Thymus. Four days post-haemorrhage. 25 A very similar cell to 24
Effect of Haemorrhage on Thymus and Bone Marrow of Starlings
465
466
M.D. Kendall
Small Lymphocytes. (Fig. 3, 6 and 14; Fig. 6, 1). Cells less than 6 pm in total diameter. The cytoplasm was scanty and normally polar in position. The nuclei were distinctly pachychromatic with a nucleolus usually visible. The cytoplasm stained slightly with basophilic dyes. These include the thymocytes of the thymus.
Larger Lymphoid Cells. (Fig. 2, 1; Fig. 3, 4, 6, 11 and 15; Fig. 6, 2, 3 and 4). A heterogeneous group of cells all of which were greater than 6.5 ~tm in diameter and had cytoplasm clearly visible round the nucleus. The nucleus was typically rounded but the degree of leptochromasia within allowed some cells to be distinguished as transitional cells (Rosse, 1976), others as medium or large lymphocytes and yet other as possible blast cells. Transitional cells had a small amount of cytoplasm that was normally basophilic but the degree of basophilia varied in different cells. The nucleus was clearly leptochromatic in large cells and less so in smaller cells. One or two nucleoli were normally present. Medium and large lymphocytes were similar in size to the transitional cells but the nucleus was always to some extent pachychromatic, and nucleoli were readily discernable. The cytoplasm was normally weakly basophilic and usually more extensive than in transitional cells. Occasional strongly acidophilic granules were observed in the cytoplasm. Possible blast cells had more cytoplasm than either lymphocytes or transitional cells. The nucleus was rounded and generally smaller than medium or large lymphocytes (typically approximately 6 pm diameter). There was one clear, often centrally placed nucleolus and sometimes possibly one other. Cells containing more than two nucleoli were regarded as medium or large lymphocytes. The nucleolus was ring-shaped and comparatively large. Nuclear chromatin was often aggregated peripherally as shown in Fig. 6, 5, but in some instances this aggregation was incomplete so that the nucleus appeared leptochromatic in part. The more leptochromatic the nucleus the more acidophilic it appeared. The cytoplasm was normally visible all round the nucleus and was slightly basophilic. These possible blast cells formed an unbroken spectrum of cells from a small dense pachychromatic cell with little cytoplasm (a small lymphocyte) to the acidophilic erythroid line described later. To facilitate quantitative observations on cell populations, all cells in this series with basophilic cytoplasm were classified as lymphoid, and if there was no cytoplasmic basophilia visible then the cells were classified as erythroid. The complete spectrum of cells is shown in (Fig. 6). Cells resembling Fig. 6, 7 were occasionally observed as doublets immediately after mitosis.
Erythroid Cells The developmental sequence of erythroid cells in chicken bone marrow, according to Lucas and Jamroz (1961), involves a large rounded basophilic cell, the erythroblast, that is progressively reduced in size whilst still retaining the rounded leptochromatic nucleus and small amount of basophilic cytoplasm until there is a hint of haemoglobin formation in the cytoplasm. At this stage the cell (that may
Effect of Haemorrhage on Thymus and Bone Marrow of Starlings
467
measure about 10 ~tm) is termed a polychromatic erythrocyte. Their drawings show very little basophilia of the cytoplasm at this stage. When the full haemoglobin colour is visible in the cytoplasm the nucleus has often become more oval. At this stage too the typical oval cell outline is formed and further reduction in nuclear size and degree of leptochromasia leads to the formation of the mature form. The thrombocytic line of development in chickens is depicted as differing from the erythrocyte line in that the initial basophilia of the cytoplasm is greater and persists for longer, the nucleus has a different (punctate) chromatin pattern although it is still leptochromatic and no haemoglobin develops. The nucleus of the mature thrombocyte is larger than that of the erythrocyte, and is a violet purple colour rather than a bluish colour with 'Romanowsky-type' stains. Cytoplasmic granules persist and are regarded as characteristic of mature circulating thrombocytes. Detailed examination of the starling smears prepared in this study showed differences that were not reconcilable with the above classical picture for developing avian erythrocytes or thrombocytes. Firstly the basophilic erythroblast cells of Lucas and Jamroz (1961) were easily observed but the degree of cytoplasmic basophilia was very intense and persisted for a long time after haemoglobinisation was observed. (This may be of little consequence as it is known that the timing of the development of haemoglobin in the erythroid cells can vary according to different physiological conditions). Secondly haemoglobin was seen to develop in the cytoplasm of cells that contained no basophilia whatsoever, were already ovoid in shape, had larger acidophilic nuclei than the erythroid cells, and were in all respects, except for a lack of cytoplasmic inclusions, thrombocytes. Thirdly, these "thrombocytes" develop from precursor cells that were only very faintly basophilic in the largest most primitive stage. They rapidly lost any basophilia so that the cytoplasm was very difficult to observe, even under phase contrast, and had throughout a very acidophilic nucleus. Small azurophilic granules were sometimes observed with Giemsa staining. Acidophilia was retained even when the cytoplasm appeared fully haemoglobinised, but it must have been lost eventually as all erythrocytes looked the same in circulating peripheral blood. Thus in this study the erythroid cells have been regarded as either basophilic (with basophilic cytoplasm) or acidophilic (with acidophilic nuclei). Combining the 'Romanowsky-type' stains with Lephene's stain for haemoglobin allows the following categories to be identified.
Basophilic, Non-Haemoglobinised Cells. (Fig. 3, 7, 8 and 12). These ranged in size from very large cells (10 Ixm diameter) to cells approximately 6 ~tm in overall diameter. Both the cells and the nuclei were rounded. The nucleus was leptochromatic, occasionally with a nucleolus. The cytoplasm was distinctly basophilic. Mitotic figures were observed, particularly in the larger cells.
Basophilic, Haemoglobinised Cells. (Figs. 2, 9; Fig.'3, 17, 19, 20 and 21). In general, smaller in size than the non-haemoglobinised cells. Following haemorrhage some cells were very small indeed (4-5 ~tm). The cytoplasm ranged from slightly green in colour, with a lot of basophilia still present, though a distinctly green stage with a smaller nucleus to a more brownish colour of the cytoplasm as the haemoglobin
468
M.D. Kendall
0 ]'''''.
~
ii
li P
60
20
B1
0
Z
.............
1O~o 0
2
4
,~ A2
~
~EO
6
8
[
DAYS
Fig. 4. Results of the differential counts performed on bone marrow smears. The lymphoid compartment (top). L-L Larger lymphoid cells (medium and large lymphocytes, transitionals and possible blast cells excluding basophilic erythroblasts); S-L small lymphocytes; P pyknotic cells. The erythroid compartment (higher middle). A1 Acidophilic erythroid cells; A2 more mature acidophilic erythroid cells; B1 Basophilic erythroid cells; B2 more mature basophilic erythroid cells (often green in colour because of some degree of Lephene reaction). Damaged cells (D) and miscellaneous cells (M) miscellaneous cells include macrophages, promyelocytes and plasma ceils. The myeloid compartment (bottom); EO eosinophils of both sizes; N Heterophils. The basophils were less than 0.5 ~o predominated. The intensity o f the brown colour was frequently reduced following haemorrhage. Mitosis was c o m m o n l y observed. The nucleus was round, and often contained acidophilic granules in the nucleoplasm. Lephene positive material was observed between the heterochromatin, sometimes giving a stronger reaction than in the cytoplasm. W h e n basophilia o f the cytoplasm was largely lost, Lephene + ve material predominated then these cells were regarded as reticulocytes. Acidophilic, Non Haemoglobinised Cells(Fig. 2, 3; Fig. 3, 16 and 23; Fig. 6, 6 and 7). The distinguishing feature o f these cells was the acidophilic nucleus, that had very
Effect of Haemorrhage on Thymus and Bone Marrow of Starlings
469
S-L
40
Z2O
............
-
.
-
'~ L - L
~P
40
Z20 AI A2
Z 2~ ...........
0
i ....
2
. . . . . . . . . ~. . . . . . . . . . . .
4
n. . . . . . . . . . . .
6
=M
8
DAYS
Fig. 5. Results of the differential counts performed on thymus smears. Abbreviations as for Fig. 4. The percentages of myeloid cells was very low and similar in all counts
little differentiation in all but the larger, presumably more primitive, cells. A pale nucleolus was sometimes seen. The cytoplasm was usually unstained and very difficult to distinguish even under phase contrast microscopy as it rarely contained any inclusions. Mitotic figures were observed. Some of the cells appeared very fragile and it was not all uncommon to find cells in which the cytoplasm was partially unstained and partially acidophilic as though nucleoplasm had streamed out from the nucleus. Lucas and Jamroz (1961) show a figure of a similar but vacuolated cell that they regard as a technical artifact. This may be so, but it is a useful occurrence as it enables the oval cell outline to be observed at least in part of the cell. Occasionally the unstained part was either faintly basophilic or more commonly, slightly 'Lephene-positive'. These acidophilic cells have been designated A1 cells in Figs. 4 and 5. A more mature form, described below was designated A2.
470
M.D. Kendall
5
5 pro,
Fig. 6. A selection of drawings of cells from the thymus 4 days post-haemorrhage. 1 Small pachychromatic lymphocyte; 2 Medium lymphocyte with a less pachychromatic nucleus (early small transitional); 3 Transitional cell. Occasionally transitional cells had very basophilic cytoplasm and then they resembled basophilic erythroblasts; 4 Possible blast cell of the acidophilic erythroid lineage. The amount of cytoplasm is variable; 5 Similar to 4 sometimes but the slightly basophilic cytoplasm can become "clear" in places; 6 Acidophilic erythroid cell with clear cytoplasm and a slight Lephene + ve reaction. The inclusions at the top are only sometimes present and they range in number from 3-12; 7 Cells with acidophilic nuclei and clear cytoplasm that are either in mitosis or close to it. No cytoplasmic inclusions or basophilia are apparent
The r o u n d e d more primitive cells appeared to be linked morphologically with large l y m p h o i d cells clearly seen 4--6 days after h a e m o r r h a g e in both the thymus and in the bone marrow. (See l y m p h o i d cell descriptions.) Smaller, m o r e dense, but still acidophilic nuclei were associated with an oval cell shape, similar to that o f an erythrocyte (A2 cells). N o r m a l l y these cells lacked any inclusions but otherwise would have been regarded as thrombocytes. Frequently, however, there was a very slight positive reaction to Lephene's stain for haemoglobin.
Acidophilic, Haemoglobinised Cells. (Fig. 2, 5; Fig. 3, 22, 24 and 25). The acidophilic fairly h o m o g e n o u s , oval nucleus can be clearly seen within the 'Lephene-positive' cytoplasm. N o inclusions or basophilia were observed in the cytoplasm. These cells were included in the counts o f reticulocytes (with the basophilic lineage reticulocytes described above).
Thromboeytes. Previous authors (Lucas and Jamroz, 1961; M u k k u r and Bradley, 1967; Santamarina, 1964) have found that thrombocytes can be confused with either p o o r l y developed erythrocytes or lymphocytes. The use o f P.A.S. to stain glycogen in the cytoplasm (Mills Westerman, 1974) m a y help with the identification.
Effect of Haemorrhage on Thymus and Bone Marrow of Starlings
471
Table 2. To show the percentage of mature erythrocytes, reticulocytes and thrombocytes that contained cytoplasmic granules. Mature erythrocytes should be Lephene + ve and not contain granules whereas thrombocytes should be Lephene - v e and contain cytoplasmic granules Day
Lephene + ve No granules
Granules
Lephene + ve+ Basophilic (green)
Lephene - ve No granules
Granules
0
61.5_+ 7.1
16.5_+ 5.4
10.2+10.0
11.7_+10.2
0.2+0.3
2
78.3_+ 3.4
7.8-+ 5.6
7.2_+ 6.2
6,7_+ 1.0
4
34.2_+28.7
1.5_+ 2.6
0.8_+ 1.4
63,3_+30.6
6
14.8_+ 7.1
13.8_+ 11.7
64.3_+ 6.4
7.0__+ 10.8
0
8
76.7-+ 14.5
4.3_+ 3.9
14.2_+20.1
4.0+ 2.8
0.8-+ 1.1
0
66.5-+ 7.5
22.3-+ 4.9
0
10.5-+ 9.8
0.7-+0.8
2
86.5-+11.6
5.3_+ 4.3
0
7.8+ 7.6
0.2+0.3
4
44.7_+ 9.5
2.3+ 0.8
0
52.3+10.0
0.7-+0.6
Bone marrow
0 0.2+0.3
Thymus
*
~
***
6
73.3_+ 9.5
17.0-+ 9.5
0
9.7+ 1.5
0
8
83.5_+ 4.9
12.5_+ 7.8
0
3.5_+ 3.5
0.5_+0.7
93.8_+ 4.7
6.1+ 4.7
0
0
0
90.6_+ 15.9
8.5_+ 13.0
0
0
0
Peripheral blood 0 All posthaemorrhage. *
P=
