Comparative Biochemistry and Physiology Part A 125 (2000) 75 – 83
www.elsevier.com/locate/cbpa
Spectral imaging of red blood cells in experimental anemia of Cyprinus carpio Chana Rothmann, Tova Levinshal, Bracha Timan, Ramy R. Avtalion, Zvi Malik * Faculty of Life Sciences, Bar-Ilan Uni6ersity, Ramat-Gan 52900, Israel Received 29 June 1999; received in revised form 9 September 1999; accepted 22 September 1999
Abstract In the present work we have studied the effect of experimental anemia induced at both low and optimal temperatures on erythropoiesis in Cyprinus carpio. The results showed that hemoglobin concentration per cell was similar in both temperature conditions, however, red blood cell (RBC) concentration was higher at the optimal temperature. Induced anemia caused an abrupt decrease in RBC concentration, while the hemoglobin concentration per cell remained unchanged. Recovery, as shown by electron microscopy, was characterized by the release of differentiating young and intermediate cells to the peripheral blood. It was revealed that with the progression of differentiation the nucleus/cytoplasm ratio decreases, the chromatin condenses and the shape of the nucleus changes from round to elliptical. Spectral imaging revealed an increase in the optical density of chromatin with the maturation of the cells. The chromatin that was dispersed over the nuclear volume in the young cells becomes highly ordered in the mature cells. Spectral similarity mapping revealed the formation of a novel structure of high symmetry, representing chromatin rearrangement during the process of cellular differentiation. © 2000 Elsevier Science Inc. All rights reserved. Keywords: Anemia; Cyprinus carpio; Differentiation; Electron microscopy; Erythropoiesis; Optical density; Spectral imaging; Temperature
1. Introduction The evolutionary process dictated a number of dramatic changes in the course of erythropoiesis. In mammalians, one of those changes is the extraction of the small pyknotic nucleus and the yielding of a mature, unnucleated erythrocyte. In contrast, in the terminal stage of the erythropoiesis in fish, the nucleus is not extracted from the cell but remains intact. In addition, the distinct living environment of the fish dictates sur-
* Corresponding author. Tel.: + 972-3-5318204; fax: +9723-5345878. E-mail address:
[email protected] (Z. Malik)
vival conditions that affect the erythropoietic process. The carp, as an ectothermic vertebrate, is seasonally adapted to temperatures ranging from 4 to 35°C. From winter to summer, standard oxygen consumption of carp rises continuously while oxygen availability in water drops gradually and may cause respiratory stress (Collazos et al., 1994). Studies point to teleostean erythrocyte as a dynamic system responsive to changes in oxygen demand and/or availability through adjustment of blood O2 carrying capacity and hemoglobin O2 affinity. In large measure these responses center upon alterations in red cell levels of organophosphate and inorganic electrolytes directly or indirectly affecting affinity (e.g. ATP, GTP, H+, Cl−,
1095-6433/00/$ - see front matter © 2000 Elsevier Science Inc. All rights reserved. PII: S 1 0 9 5 - 6 4 3 3 ( 9 9 ) 0 0 1 5 7 - 9
76
C. Rothmann et al. / Comparati6e Biochemistry and Physiology, Part A 125 (2000) 75–83
Mg2 + ) as well as changes in hemoglobin abundance (Houston and Murad, 1995). Hematological response in fishes embodies two important limitations: 1. Erythrocyte maturation is accompanied by the elimination of the mitochondrial, ribosomal, and hemosomal complements, the organelles associated with heme and globin synthesis and assembly (Sekhon and Beams, 1969; Keen et al., 1989; Speckner et al., 1989). 2. Mature cells do not take up 55Fe and do not appear to synthesize hemoglobin (Murad et al., 1993). As a result, the isomorphic profiles of individual erythrocytes are fixed during maturation (Byrne and Houston, 1988; Marinsky et al., 1990) and while remaining mitochondria could permit changes in cell modular levels, only physical processes can alter the hemoglobin present (Houston and Ropert, 1976). Thus, maturation is a crucial limiting step in the adaptive process. Furthermore, hematological response does not occur as an isolated process. O2 delivery to tissues is a function of blood O2 content and cardiac output which could compensate for deficits in O2-carrying capacity (Cameron and Davis, 1970; Wood et al., 1979). However, elevation of blood O2 tension in this way reduces hypoxemic stimulation of erythropoiesis, curtails further increase in O2-carrying capacity, and forces reliance on more metabolically costly systemic functions. Earlier studies have demonstrated a common pattern of hematological response to a variety of respiratory challenges, including transient hypoxia, a temperature-induced increase in O2 demand, hemorrhage, and experimentally induced anemia (Houston et al., 1988; Murad et al., 1990, 1993; Houston and Murad, 1991, 1992, 1995). The peripheral blood of fish in normal conditions usually contains a small proportion of immature red blood cells (RBCs), identified microscopically by their circular shape as opposed to the elliptical shape of the mature cells. Like the RBCs in all non-mammalian vertebrates, carp cells retain their nucleus and some cytoplasmic organelles while circulating in the peripheral blood (Fange, 1994). In individuals that have previously suffered respiratory stress such as hypoxia or experimentally induced anemia (bleeding or phenylhydrazine treatment) the population of immature red cells may be increased temporarily (Houston and Murad, 1995).
Standard analysis of blood cells for the determination of differentiation and pathological conditions is usually based on staining with May– Grunwald–Giemsa (MGG) or Romanowsky techniques which employ the dyes azure B and eosin. Spectroscopic selected area microanalysis has been shown to enhance the data obtained from cells stained by standard methods (Spina et al., 1992). In utilizing this method it has become apparent that the stained nucleus may preserve much more information on its fine structure than the bare eye can define by conventional light microscopy. Spectral imaging is a young technology that combines spectroscopy and imaging to provide the spectrum of light for every pixel of an image. Spectroscopy carries information on the interactions of light with matter, while imaging records and provides spatial information on the studied objects (Rothmann et al., 1998). In spectral imaging of histological and cytological specimens, \ 104 pixel-spectra (400–850 nm) are obtained for an image, providing more information than conventional gray scale image analysis (Stenkvist et al., 1978), even when the latter is used in combination with color filters (Wells et al., 1992). Spectral imaging differentiates between closely related colors even when the total intensity is similar and is thus very sensitive and specific. The aim of the present study was to determine the effect of experimentally induced anemia at both low and optimal temperatures (14 and 25°C, respectively) on the ultrastructure of the erythroblasts and erythropoiesis, and to characterize the nuclear changes using spectral imaging in comparison to transmission and scanning electron microscopy.
2. Materials and methods
2.1. Fish and tanks Carp (Cyprinus carpio) of 250–300 g in weight were kept in fresh water in a density of 1 kg body weight/50 l. The tanks were equipped with controlled warming and cooling devices. The water were continuously aerated, and exchanged twice a week. The fish received dry fish food containing 20% protein at a rate of 1–2% of their body weight. The fish were individually marked as previously reported (Avtalion et al., 1973). Constant temperature tanks were kept at 14 or 2591°C.
C. Rothmann et al. / Comparati6e Biochemistry and Physiology, Part A 125 (2000) 75–83
2.2. Bleeding and induction of anemia Bleeding was performed by puncture of the caudal vessels. For hematological determinations, 0.5 ml blood was collected in a heparinized tube. Mild anemia was induced by bleeding of 4 ml at the beginning of each experiment. This quantity presents about 30% of the total blood volume of the carp used in these experiments (Avtalion et al., 1974).
2.3. Determination of hematological parameters The heparinized blood was washed in a 0.9% NaCl solution, centrifuged and RBCs were counted by a hemocytometer. To estimate hemoglobin concentration per 105 cells, the washed RBCs were lysed in 0.005 M phosphate buffer pH 6.8. Samples were diluted in modified Drabkin solution (potassium ferricyanide 0.006 mM, potassium cyanide 0.8 mM in phosphate buffer 1 mM) which converted the hemoglobin to cyanmethemoglobin. A standard curve was plotted according to known hemoglobin solutions at 545 nm in a Gilford 250 spectrophotometer, and hemoglobin concentration in the samples was calculated.
2.4. Preparation for transmission electron microscopy RBCs were fixed by 2.5% glutaraldehyde in phosphate buffer pH 7.2, post-fixed with 2% osmium tetroxide, embedded in Epon 812, thin sectioned by a LKB Ultratome III, and stained with uranyl acetate and lead citrate. The samples were examined using a Jeol 1200EX transmission electron microscope.
2.5. Preparation for scanning electron microscopy Cells were fixed by 2.5% glutaraldehyde in phosphate buffer pH 7.2, then washed in the same buffer, and post fixed by 2% osmium tetroxide. The third step of fixation was performed with a solution of tannic acid-guanidine hydrochloride. The triple-fixed cells were dehydrated in graded alcohol solutions and then the alcohol was exchanged for Freon-112 by graded Freon solutions. The cells were air-dried, gold coated and examined by a Jeol 840 scanning electron microscope.
77
2.6. Fourier-transform multipixel spectrometry system for microscopy Spectral image analysis of MGG-stained RBCs was performed using the SpectraCube SD-200 (Applied Spectral Imaging, Migdal HaEmek, Israel) combined with a bright field microscope. The SpectraCube system consists of an interferometer and a CCD camera. During a measurement (20 s), each pixel of the CCD (512× 512) is collecting the interferogram, which is then Fourier-transformed to give the spectrum (Malik et al., 1996). As a result, spectral imaging acquires a so-called cube whose appellate signifies the two spatial dimensions of a flat sample (x and y), and the third spectrum dimension representing light intensity at any wavelength. The calculated pixel size in a spectral image is 0.04 mm2. The spectral resolution (FWHM, full width at half maximum) is 5 nm at 400 nm (12 nm at 600 nm) and the spectral range (\5% response) is 400–1000 nm (Garini et al., 1996).
2.7. Spectral similarity mapping Similarity mapping is useful in a situation in which the sample is composed of a number of spatially separated components, each characterized by a known and unique spectrum, and the task is to detect and map all components. The steps of this algorithm were as follows: 1. The spectra of the distinct nuclear regions were stored in a ‘spectral library’. 2. For every pixel of the cube, a comparison was made between its measured spectrum and all the spectra of the library. 3. Each pixel in the image was identified with the most similar spectrum in the library. 4. Each pixel was displayed in a previously established color identifying the specific library spectrum, forming a so called ‘classified image’. The comparison formula for the second step of the similarity mapping algorithm was as follows — n functions f nx,y (n is the number of spectra in the library) are defined for every pixel of spatial coordinates x and y as follows: fx,y =
&
l2
l1
[Ix,y (l)− I0(l)]2dl
1/2 1/2
(1)
where the integral over l stands for an integral over a predetermined spectral range l1 –l2, Ix,y (l)
78
C. Rothmann et al. / Comparati6e Biochemistry and Physiology, Part A 125 (2000) 75–83
is the spectrum of the pixel in question, and In (l) are the n spectra of the library. Associated with each of these spectra are n artificial colors. In this analysis scheme each pixel is displayed in the color which corresponds to the library spectrum for which Eq. (1) is minimum. The resulting ‘classified image’ reveals areas with similar chromophore compositions. In some cases additional quantitative assessments of the classified regions’ size is needed, either in absolute or relative terms. Gray shades can be added according to the value of the selected f nx,y. When there is only one component (n =1) with which to compare each pixel, a black and white image is all that is needed, and the gray levels according to the values of f 1x,y are used for a good rendition of the image. The presence of the component in question is then easily seen in the bright pixels, whereas the dark ones represent its absence, and the gray levels may correspond to different amounts of the component.
3. Results The effect of constant acclimation on erythropoiesis was studied by randomly dividing 20 carp from breeding pools into two groups, ten in each, which were kept for 30 days at 14 and 25°C, respectively. Anemia was induced by withdrawal of 4 ml blood from each fish and a renewal of the RBC
pool was expected after this operation. RBC number and hemoglobin concentration were determined at the time of anemia induction, and after 15 and 35 days. The RBC concentration was significantly lower in the fish held at 14°C than in those held at 25°C (18.37 and 23.93 cells/ml, respectively) before bleeding, and decreased at day 15 after the induction of anemia to 5.8 9 0.15×108 and 10.99 90.1× 108 cells/ml, respectively (Fig. 1A). Thirty-five days later, a significant increase in the number of cells could be observed. Hemoglobin concentrations in the cells at 0 and 35 days after bleeding were similar in fish held at 14 and 25°C (Fig. 1B). The morphology of RBCs from the two groups of fish, normal and anemic, was analyzed by scanning and transmission electron microscopy. Fig. 2A presents the normal RBC population from a high temperature acclimated fish. Three distinct RBC populations of large, intermediate and small longitude diameter (\ 20, 16–20 and B 16 mm, respectively) could be seen in the anemic fish (Fig. 2B). Transmission electron microscopy revealed a population of early to matured stages of the erythroid lineage. The early erythroblasts were poorly differentiated cells and their cytoplasm contained organelles and was not hemoglobinized (Fig. 2C). The intermediate cells were characterized by a decrease in nuclear dimensions, nuclear shrinkage, a variable number of ribosomes-polyribosomes, elimination of cellular organelles and pronounced hemoglobinization
Fig. 1. The implications of temperature level on the response to experimental anemia and the recovery process. (A) RBC concentration and (B) hemoglobin concentration per cell were measured in carps at 14 and 25°C. Each column is an average 9 S.E. of three fish.
C. Rothmann et al. / Comparati6e Biochemistry and Physiology, Part A 125 (2000) 75–83
79
Fig. 2. Characterization of peripheral RBCs by electron microscopy. A scanning electron micrograph of (A) normal blood and of (B) peripheral blood of anemic carp 7 days after bleeding. (Original magnification × 1200 bar = 10 mm). A transmission electron micrograph of (C) early erythroblast (magnification ×16 000) (D) intermediate cell (magnification ×12 000) and (E) mature RBCs (magnification× 16 000).
Fig. 3. The percentage of RBCs — early erythroblast, intermediate and mature was determined following bleeding at 14 and 25°C.
(Fig. 2D). The matured RBC was fully hemoglobinized, the number of ribosomes depleted, only remnants of mitochondria were found, the nucleus was maximally shrunken and the nuclear envelope space was pronounced (Fig. 2E). Fig. 3A and B show the kinetics of development of these three populations at 14 or 25°C. The immediate effect of bleeding was a reduction in the percentage of both the large (mature) and
intermediate cells with a parallel in the small size cells (early erythroblasts). Later on, a progressive increase in large cells appeared, while the number of the small and intermediate cells underwent a graduate decrease (Fig. 3A, B). These erythroid developmental stages were investigated using multi-pixel spectral analysis of the red cells stained by MGG (Rothmann et al., 1997). The spectra revealed two nuclear domains,
80
C. Rothmann et al. / Comparati6e Biochemistry and Physiology, Part A 125 (2000) 75–83
a chromatin domain with low intensity light transmittance and high optical density, and an interchromatin domain with high intensity light transmittance and low optical density. Each spectrum in Fig. 4A–C is the average of a sample of 100 pixels arbitrarily chosen from both domains of ten cells. Spectral similarity mapping using a single reference spectrum from each region revealed the organization of the two nuclear components in the carp erythrocytes. By using a reference interchromatin spectrum for a first similarity mapping (Fig. 5A – C) and a reference chromatin spectrum for a second (Fig. 5D – F), it was revealed that the major nuclear area of the young erythroblasts is composed of chromatin regions arranged in small unconnected patches (Fig. 5A, D). With the progression of the differentiation process, the number of chromatin patches is de-
Fig. 5. MGG stained RBCs were analyzed by spectral imaging. Spectral similarity mapping was performed for the three stages of development: (A, D) early erythroblast (B, E) intermediate and (C, F) mature RBC, using the low intensity light transmittance spectrum as a reference spectrum for the first mapping (A – C) and the high intensity light transmittance spectrum as a reference spectrum for the second mapping (D – F), bar=5.7 mm.
creased and a symmetrical circular arrangement can be observed (Fig. 5B, E). The mature cells exhibited a highly ordered distribution of chromatin with a central chromatin spot (Fig. 5C, F). Table 1 shows the values of interchromatin and chromatin distribution in differentiating erythroblasts. The chromatin values depict a trend of chromatin condensation, which starts at 77.6% of total nuclear area in young erythroblasts and reaches 34.28% in the mature cells.
Fig. 4. Optical density spectrum of the low and high optical density domains of the nucleus. Each spectrum is the average of a sample of 100 pixels arbitrarily chosen from the two domains of ten cells. (A) Young erythroblast (B) intermediate and (C) mature RBC.
4. Discussion In the present study we have used two experimental systems, temperature changes and anemia,
C. Rothmann et al. / Comparati6e Biochemistry and Physiology, Part A 125 (2000) 75–83
81
any doubt (Haaf and Schmid, 1991; Rothmann et al., 1997). It has been proposed that repetitive sequences act as a structural center for the extension and condensation of chromatin (Manuelidis and Borden, 1988). Thus, compartmentalization of chromatin could provide a structural framework for efficient processing of nuclear events (Vourc’h et al., 1993). Based on this hypothesis, the specific nature of the compartmentalization could reflect the physiological state of a given cell (Emmerich et al., 1989; Popp et al., 1990; Van Dekken et al., 1990; Vourc’h et al., 1993). However, it should be emphasized that a universally valid principle of chromosome arrangement does not exist (Haaf and Schmid, 1991). Recently, Cremer et al. (1993) proposed a model predicting that the surface of chromosome territories and a space formed between them provide a network-like three-dimensional nuclear compartment for gene expression, mRNA splicing and transport, termed the interchromosome domain (ICD) compartment (Zirbel et al., 1993). On the basis of the present results, we speculate that the circular and windmill patterns revealed by spectrally resolved imaging define a three-dimensional compartmentalization of chromatin in the differentiating nucleus. The low optical density regions creating the windmill pattern may represent the ICD compartment of Zirbel et al. (1993) while the high optical density regions represent the chromatin. According to the model proposed by Cremer et al. (1993), chromatin loops with genes that are permanently or intermittently expressed in a given cell type should be located at or close to the surface area of each chromosome territory. This arrangement enables RNA transcripts to be directly released into the ICD compartment, go through processing in a topologically, highly ordered manner and be transported to the nuclear pores (Carter et al., 1993; Xing et al., 1993). A model of genome
in order to characterize the distinct stages of carp erythropoiesis. Fish and human erythroblasts share common processes of maturation of the cytoplasm and nucleus. However, in contrast to human erythropoiesis, the mature RBCs of fish retain their nucleus. The significant decrease in both cell number and hemoglobin obtained at 14°C results from the fact that at a low temperature, the solubility coefficient of O2 is high and allows high carrying capacity levels. Elevation in environmental temperature decreases the solubility of O2 and lowers the carrying capacity of RBCs (not shown) resulting in an increased number of RBCs to provide more binding sites for O2. The adaptation of the erythropoietic system to distinct temperature levels did not affect the hemoglobin concentration per cell. The induction of anemia caused a rapid decrease in RBC concentration, which was dealt with by the erythropoietic system by releasing young and intermediate cells. The hemoglobin concentration per cell remained unchanged, as observed by electron microscopy. The young cells appeared small (B16 mm) and round with a large nucleus and chromatin regions arranged in small unconnected patches. Intermediate cells appeared larger (16–20 mm) and their nucleus appeared smaller and more condensed. Thus, with the progression of differentiation, the nucleus/cytoplasm ratio decreases, the chromatin condenses and the shape of the nucleus changes from round to elliptical. The chromatin that was dispersed over the nuclear volume in the young cells becomes highly ordered in the mature cells. Spectral imaging revealed an increase in the optical density of chromatin of mature cells and a highly symmetrical circular arrangement with a central chromatin spot. The existence of highly ordered organizational patterns in the cell nucleus appears to be beyond Table 1 The relative area of distinct nuclear domainsa Cells stage
Nuclear area (mm2)
Interchromatin (%)
Euchromatin (%)
Young Intermediate Mature
39.391.0 32.89 3.6 14.0 91.0
22.4 28 65.7
77.6 71.95 34.28
a The relative area of chromatin and interchromatin as measured in young erythroblasts, intermediate and mature cells. Each value is the average of ten nuclei +6 S.E.
82
C. Rothmann et al. / Comparati6e Biochemistry and Physiology, Part A 125 (2000) 75–83
architecture in human sperm cells was presented by Zalensky et al. (1995) positioning chromosome telomeres in the periphery of the nucleus, while centromeres were located in the center, which supports Cremer’s hypothesis. The symmetry observed in the nuclei could be maintained by electric forces. According to Cremer et al. (1993), short range (nm distances) and long range (mm distances) electric forces due to charge distribution effects of chromosome territories and other nuclear components may be involved in the maintenance of the ICD compartment (Cremer et al., 1993).
Acknowledgements This study was supported by a grant of Applied Spectral Imaging, Migdal HaEmek, Israel. We gratefully thank Ms Judith Hanania for her help in editing the manuscript and for technical assistance and Mr Jacob Langsam for his skilful assistance. The experiments comply with the current laws of Israel.
References Avtalion, R.R., Wojdani, Z., Malik, Z., Shahrabani, R., Ducziminer, M., 1973. Influence of environmental temperature on the immune response in fish. Curr. Top. Microbiol. Immunol. 61, 1–35. Avtalion, R.R., Gorlin, A., Gutwirth, E., Wojdani, A., 1974. Determination of blood volume in fish, a new method. Bamidgeh Bull. Fish. Cult. Isr. 26, 16–20. Byrne, A.P., Houston, A.H., 1988. Use of phenylhydrazine in the detection of responsive changes in hemoglobin isomorph abundances. Can. J. Zool. 66, 758–762. Cameron, J.N., Davis, J.C., 1970. Gas exchange in rainbow trout (Salmo gairdneri ) with varying blood oxygen capacity. J. Fish. Res. Board. Can. 27, 1069– 1085. Carter, K.C., Bowman, D., Carrington, W., et al., 1993. A three dimensional view of precursor messenger RNA metabolism within the mammalian nucleus. Science 259, 1330–1335. Collazos, M.E., Ortega, E., Barriga, C., 1994. Effect of temperature on the immune system of a cyprinid fish (Tinca tinca L.) — blood phagocyte function at low temperature. Fish Shellfish Immunol. 4, 231–238. Cremer, T., Kurtz, A., Zirbel, R., et al., 1993. Role of chromosome territories in the functional compart.
mentalization of the cell nucleus. Cold Spring Harb. Symp. Quant. Biol. 58, 777 – 792. Emmerich, P., Loos, P., Jauch, A., et al., 1989. Double in situ hybridization in combination with digital image analysis — new approach to study interphase chromosome topography. Exp. Cell. Res. 181, 126 – 140. Fange, R., 1994. Blood cells, haemopoiesis and lymphomyeloid tissues in fish. Fish Shellfish Immunol. 4, 405 – 411. Garini, Y., Katzir, N., Cabib, D., Buckwald, R.A., Soenksen, D., Malik, Z., 1996. Spectral bio-imaging. In: Wang, X.F., Herman, B. (Eds.), Fluorescence Imaging Spectroscopy and Microscopy. Wiley, New York, pp. 87 – 124. Haaf, T., Schmid, M., 1991. Chromosome topology in mammalian interphase nuclei. Exp. Cell. Res. 192, 325 – 332. Houston, A.H., Ropert, R., 1976. Immediate response of the hemoglobin system of the goldfish, Carassius auratus, to temperature change. Can. J. Zool. 54, 1737 – 1741. Houston, A.H., Murad, A., Gray, J.D., 1988. Induction of anemia in goldfish, Carassius auratus, by immersion in phenylhydrazine hydrochloride. Can. J. Zool. 66, 729 – 736. Houston, A.H., Murad, A., 1991. Hematological characterization of goldfish, Carassius auratus L., by image analysis; effects of thermal acclimation and heat shock. Can. J. Zool. 69, 2041 – 2047. Houston, A.H., Murad, A., 1992. Erythrodynamics in goldfish, Carassius auratus L. — temperature effects. Physiol. Zool. 65, 55 – 76. Houston, A.H., Murad, A., 1995. Erythrodynamics in fish — recovery of the goldfish Carassius auratus from acute anemia. Can. J. Zool. 73, 411 – 418. Keen, J.E., Calarco Steel, A.M., Houston, A.H., 1989. The circulating erythrocytes of the rainbow trout (Salmo gairdneri ). Comp. Biochem. Physiol. A. 94, 699 – 711. Malik, Z., Cabib, D., Buckwald, R.A., Talmi, Y., Garini, Y., Lipson, S.G., 1996. Fourier-transform multipixel spectroscopy for quantitative cytology. J. Micros. 182, 133 – 140. Manuelidis, L., Borden, J., 1988. Reproducible compartmentalization of individual chromosome domains in human CNS cells revealed by in situ hybridization and three-dimensional reconstruction. Chromosoma 96, 397 – 410. Marinsky, C.A., Houston, A.H., Murad, A., 1990. Effect of hypoxia on hemoglobin isomorph abundances in rainbow trout, Salmo gairdneri. Can. J. Zool. 68, 884 – 888. Murad, A., Houston, A.H., Samson, L., 1990. Hematological response to reduced O2-carrying capacity, increased temperature and hypoxia in Carassius auratus L. J. Fish Biol. 36, 289 – 305.
C. Rothmann et al. / Comparati6e Biochemistry and Physiology, Part A 125 (2000) 75–83
Murad, A., Everill, S., Houston, A.H., 1993. Division of goldfish erythrocytes in circulation. Can. J. Zool. 71, 2190–2198. Popp, S., Scholl, H.P., Loos, P., et al., 1990. Distribution of chromosome 18 and X centric heterochromatin in the interphase nucleus of cultured human cells. Exp. Cell. Res. 189, 1–12. Rothmann, C., Cohen, A.M., Malik, Z., 1997. Chromatin condensation in erythropoiesis resolved by multi-pixel spectral imaging — differentiation versus apoptosis. J. Histochem. Cytochem. 45, 1097–1108. Rothmann, C., Bar-Am, I., Malik, Z., 1998. Review — spectral imaging for quantitative histology and cytogenetics. Histol. Histopathol. 13, 921–926. Sekhon, S.S., Beams, H.W., 1969. Fine structure of the developing trout erythrocyte and thrombocyte with special reference to the marginal band and cytoplasmic organelles. Am. J. Anat. 125, 353–373. Speckner, W., Schindler, J.F., Albers, C., 1989. Age-dependent changes in volume and haemoglobin content of erythrocytes in the carp (Cyprinus carpio L.). J. Exp. Biol. 141, 133–149. Spina, D., Disanto, A., Luzi, P., et al., 1992. Novel, contrast gradient-oriented, automated chromatin texture analysis — feasibility study on nuclei from benign and malignant breast epithelial cell lines in fine needle aspirates. Virchows Arch. B. Cell. Pathol. 62, 119–124. Stenkvist, B., Westman-Naeser, S., Vegelius, J., et al., 1978. Computerized nuclear morphology as an objective method for characterizing human cancer cell populations. Cancer Res. 38, 4688–4977.
.
83
Van Dekken, H., Arkesteijn, G.J., Visser, J.W., Bauman, J.G., 1990. Flow cytometric quantification of human chromosome specific repetitive DNA sequences by single and bicolor fluorescent in situ hybridization to lymphocyte interphase nuclei. Cytometry 11, 153 – 164. Vourc’h, C., Taruscio, D., Boyle, A.N., Ward, D.C., 1993. Cell cycle-dependent distribution of telomeres, centromeres, and chromosome-specific subsatellite domains in the interphase nucleus of mouse lymphocytes. Exp. Cell. Res. 205, 142 – 151. Wells, W.A., Rainer, R.O., Memoli, V.A., 1992. Basic principles of image processing. Am. J. Clin. Pathol. 98, 493 – 501. Wood, C.M., McMahon, B.R., McDonald, D.G., 1979. Respiratory, ventilatory and cardiovascular responses to experimental anaemia in the starry flounder, Platichthys stellatus. J. Exp. Biol. 82, 139 – 162. Xing, Y., Johnson, C.V., Dobner, P.R., Lawrence, J.B., 1993. Higher level organization of individual gene transcription and RNA splicing. Science 259, 1326 – 1330. Zalensky, A.O., Allen, M.J., Kobayashi, A., Zalenskaya, I.A., Balhorn, R., Bradbury, E.M., 1995. Welldefined genome architecture in the human sperm nucleus. Chromosoma 103, 577 – 590. Zirbel, R.M., Mathieu, U.R., Kurz, A., Cremer, T., Lichter, P., 1993. Evidence for a nuclear compartment of transcription and slicing located at chromosome domain boundaries. Chromosom. Res. 1, 93 – 106.
