| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
INNOVATIVE METHODOLOGY
1Department of Physiology, University of Bergen, 5009 Bergen; and 2Institute for Nutrition Research, University of Oslo, 0316 Oslo, Norway
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
40-kDa molecular mass range. Western blotting of the cytokines erythropoietin and granulocyte colony-stimulating factor revealed generally higher amounts in the centrifugate than in the plasma. This difference was augmented during increased hematopoietic activity induced by inflammation or bleeding in rats. We conclude that the centrifugation method provides representative samples of bone marrow extracellular fluid and that extracellular signaling responses to altered hematopoiesis are more clearly reflected locally in the bone marrow interstitium than in plasma.
interstitial fluid; cytokines; colloid osmotic pressure; inflammation; humans
Much work is in progress to test drugs that can modulate hematopoiesis in patients with hematological malignancies, as well as other disorders associated with reduced bone marrow function, like myelofibrosis or aplastic anemia. The therapeutic window of these drugs and thus the limits of what is effective and what is toxic are mostly unknown. It would therefore be of great therapeutic importance to sample extracellular fluid from the bone marrow and investigate its contents. Furthermore, recent data indicate that not only solid tumours, but also blood neoplastic disorders, are at least in part dependent on enhanced bone marrow angiogenesis for their development and progression. Several pro- and antiangiogenic molecules have been identified as possible targets for mechanism-based therapy (6, 19). Consequently, it is essential to identify and quantify these molecules locally within the bone marrow, i.e., in the marrow extracellular fluid.
The interstitium in solid tumors is also difficult to access. Recently, we showed (22) that extracellular fluid can be isolated from tumors by exposing tumor tissue to centrifugation (G forces less than 400 g). Having excluded wicks and microdialysis as alternative methods to gain access to bone marrow interstitium because of their more traumatic nature, these tumor experiments led to the hypothesis that a similar centrifugation method could also be applicable to isolate bone marrow extracellular fluid. Specifically, our aims were to study whether representative samples of extracellular fluid could be isolated from the rat tibial and femoral marrow and from human bone marrow biopsies by centrifugation. If so, we would expect that stimulation of hematopoiesis would be reflected in changed local cytokine levels. To further validate our isolation procedure, we therefore examined the levels of hematopoietic cytokines in extracellular fluid sampled during increased erythro- or leukopoiesis.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Centrifugation technique. Immediately after euthanasia, the femur and tibia were exposed and dislocated. With the use of stereomicroscopy, five to eight small cavities (diameter 200–250 µm) were carefully drilled in the bone ends until bone marrow tissue became visible. The opposite bone ends were then cut off and the prepared femur (weight range 0.20–0.32 g) and tibia (weight range 0.21–0.30 g) were transferred to 2-ml centrifuge tubes provided with a nylon mesh basket. Isolation of bone extracellular fluid was performed as detailed (1, 22). Fluid that accumulated at the bottom was collected in graduated, unheparinized glass microcapillaries for volume measurement, carefully avoiding any debris that accumulated at the bottom of the centrifuge tube. The capillaries were either closed immediately for later analysis or their contents were diluted in buffer for high-performance liquid chromatography (HPLC).
To optimize the sampling conditions, we systematically varied combinations of centrifugal force and pore size of the nylon mesh basket sample holder. In an attempt to reduce the amount of red blood cells in the centrifugate, we tested nylon meshes with pore sizes 25, 15, and 5 µm. Reducing the mesh size resulted in hemolysis in the samples, more so with the smallest pore size. Consequently, we used centrifugations with mesh pore size of 15–25 µm.
To reduce the risk of compression of cellular elements of the bone marrow, the centrifugation speed was as low as possible to collect a fluid sample. In initial experiments, the samples were spun at 1,000 rpm (106 g) for 10 min before the speed was increased successively in steps of 100 rpm until a sample appeared at the bottom of the tube. Usually, this did not occur until a speed of 1,300 rpm (179 g) was used, whereas fluid sampling was always possible at 1,500 rpm (239 g). Therefore 1,300–1,500 rpm, later described as <239 g, was used for sampling unless otherwise stated. To study the potential effect of cell compression, consecutive centrifugations using 1,700 (307 g) and 2,000 rpm (424 g) were performed, interspersed by sampling of the expelled tissue fluid. With increasing G force a higher amount of erythrocytes accumulated in the centrifugate. At centrifugation speeds of 2,500 rpm (663 g) and above, massive hemolysis and blurring of the centrifugate occurred, and accordingly these samples were discarded.
Analyses of bone marrow fluid. Our aim was to isolate fluid from the bone marrow interstitium. To estimate any contribution of intracellular fluid, we therefore studied the concentration ratio of an extracellular tracer in bone marrow fluid and plasma collected from the rats (n = 17). After anesthesia and insertion of a polyethylene (PE)-50 catheter in a jugular vein, both kidney pedicles were ligated via flank incisions, and 60–70 µCi 51Cr-labeled EDTA (Nycomed-Amersham;, Buckinghamshire, UK) in a volume of 0.2–0.4 ml was injected intravenously for distribution in the total extracellular fluid and measurement of its volume. After 120 min of equilibration (20, 24), blood was sampled by cardiac puncture, and the rat was killed. To investigate whether the equilibration period affected the tracer concentration in the centrifugate, the 51Cr-labeled EDTA circulation time was extended for up to 300 min in some experiments. The femur and tibia were removed, transferred to centrifuge tubes, and handled as described above. Bone marrow fluid isolated in these experiments was pipetted into 0.5 ml saline in counting vials.
We determined the extracellular fluid distribution volume utilizing 51Cr-labeled EDTA in 12 other rats. The bone marrow content was isolated as described (8) except that compressed air was used for flushing the bone cavity. Furthermore, for comparison and as a reference, samples taken from the bone marrow, spongious bone, and back skin for calculation of plasma equivalent distribution volumes [(cpm/g tissue)/(cpm/ml plasma), where cpm is counts/min] were placed in tared, covered vials, and weighed. Samples were counted in a LKB gamma counter (model 1282 Compugamma; Turku, Finland).
HPLC of fluid isolated from bone marrow. The distribution of macromolecules in bone marrow fluid isolated by centrifugation, elution fluid of intact or homogenized bone marrow, and plasma was determined by HPLC using Superose 12 HR 10/30 size exclusion column (Pharmacia-Biotech; Uppsala, Sweden) with optimal separation range of 10–300 kDa, as previosly described (22). The protein concentration in the elution fluid was measured by UV detection at 280 nm (Spectraseries UV10; San Jose, CA) (22).
Colloid osmotic pressure The colloid osmotic pressures (COP) in bone marrow fluid and in plasma were measured in a colloid osmometer designed for submicroliter samples (23), using membranes with a cutoff size of 10 kDa. Usually, samples of 0.5–1 µl were applied to the osmometer membrane. The coefficient of variation for this osmometer has been estimated to 3.2% for rat and 2.8% for human plasma (23).
Induction of sterile inflammation or hemorrhage. To induce a sterile peritonitis, we injected intraperitoneally a casein digest (4 ml, 12.5% in 0.9% NaCl; Bacto-Tryptone, Difco; Detroit, MI) into four awake rats. This treatment results in an approximately fivefold increased blood neutrophil number within about 5 h, mainly due to an enhanced release from the bone marrow storage (11). At this time point the rats were anesthetized, and bone marrow fluid was isolated, as described above.
Hemorrhage was induced in three other rats anesthetized with Equithesin. A PE-50 catheter was placed into one carotid artery, tunneled subcutaneously to the neck, and secured. When recovered and awake, these rats were bled 1.5% of their body weight (3.5 ml). Plasma isolated during the bleeding served as control. Four days later the rats were anesthetized, and bone marrow fluid and plasma were isolated as described above.
Human studies. We wanted to extend the animal studies to examine whether the centrifugation method for bone marrow fluid isolation was also applicable to human bone marrow. The protocol was approved by the Regional Ethics Committee for Medical Research, Health Region II in Norway, and written consent was obtained from each study subject. We studied six adult, healthy males (age 33–57 y, body weight 56–83 kg). A blood sample (5 ml) was obtained from all subjects by puncture of an antecubital vein. After infiltration of the skin and periost with local anesthesia (5 mg/ml; Xylocain, AstraZeneca; London, UK), a cylindrical bone marrow biopsy was collected from one iliac crest. The centrifugation of the collected bone marrow biopsies was performed as outlined above.
Analysis of proteins in bone marrow fluid and plasma. We used species-specific monoclonal antibodies (rat: Research Diagnostics; Flanders, NJ; human: Pharmingen; San Diego, CA) raised against either erythropoietin or granulocyte colony-stimulating factor (G-CSF) to identify these cytokines in samples of bone marrow extracellular fluid obtained from either the rats or the human subjects, and using Western blotting as previously detailed (7). Briefly, the centrifugate proteins were separated on SDS-PAGE (7.5%), then transferred onto nitrocellulose membranes, and probed with relevant antibodies, before analysis with an enhanced chemiluminescence detection kit (ECL; Amersham, Little Chalfont, UK). Nonspecific binding was blocked with 10% skimmed milk. Equal volumes (usually 10–15 µl) of the extracellular fluid and plasma samples were added to the SDS-PAGE gels.
Concentrations of erythropoietin and granulocyte colony-stimulating factor (G-CSF) in human plasma were quantified in triplicates with ELISA-kits (Quantikine, R&D Systems; Minneapolis, MN).
Statistics. Values are reported as means ± SE. Differences were tested with two-tailed t-tests or ANOVA, followed by Tukey tests as appropriate, and were assumed significant for P < 0.05.
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
5 µl) after centrifugation of the human biopsies. Tracer recovery in rat bone marrow extracellular fluid. To determine the possible contamination of intracellular fluid in the isolated samples of bone marrow extracellular fluid, we measured the recovered centrifugate tracer (51Cr-labeled EDTA) that had equilibrated in extracellular fluid in rats (Fig. 1). At a centrifugation speed of <239 g, the average concentration of 51Cr-labeled EDTA in the bone marrow fluid relative to that of the plasma was 0.85 ± 0.03 (n = 13), which was significantly different from 1.0. Increasing the G force in consecutive centrifugations resulted in lower ratios than observed at <239 g. Thus for centrifugation speeds of 307 and 424 g, ratios of 0.64 ± 0.07 (n = 5) and 0.39 ± 0.072 (n = 5) were observed, respectively, both lower than that of <239 g (P < 0.01 and 0.001, respectively). The ratios for tibia and femur did not differ significantly, and thus the results from the two tissues were pooled. In some experiments, the tracer equilibration period was extended to 180 (n = 2) and 300 min (n = 2) without any further effect on the centrifugate to plasma 51Cr-labeled EDTA ratio (data not shown).
|
The amount of bone marrow isolated in the fluid volume determination experiments averaged 0.0289 ± 0.0022 g (range 0.0201–0.0377 g, n = 12). The extracellular fluid volumes measured as the 120-min distribution volume of 51Cr-labeled EDTA was 0.12 ± 0.01 ml/g wet wt (n = 12) in the bone marrow, with corresponding volumes of 0.14 ± 0.02 and 0.43 ± 0.03 ml/g wet wt (n = 12) in spongious bone and back skin, respectively.
COP in bone marrow fluid. COP in fluid isolated from the rat tibia and femur and human bone marrow are shown in Fig. 2. In rats (n = 8), COP averaged 19.4 mmHg in bone marrow, with a corresponding pressure in plasma of 24.1 mmHg (P < 0.05). In human controls (n = 6), the bone marrow fluid COP averaged 23.2 mmHg, which was clearly lower than the corresponding pressure in plasma of 33.0 mmHg (P < 0.05). The relative amounts of albumin and IgG did not change when the G force was increased.
|
Protein patterns in bone marrow fluid analyzed with HPLC. Figure 3 shows elution patterns of fluid isolated from rats with rat plasma (Fig. 3A) as a reference, showing the elution volumes of typical plasma proteins. Figure 3B depicts the pattern for femoral bone marrow fluid isolated by centrifugation at 239 g in control conditions. We observed that albumin dominated quantitatively and that IgG constituted a minor fraction of proteins in the eluent. The relative amounts of albumin and IgG did not change when the G force was increased. For molecules smaller than albumin, the elution pattern of bone marrow fluid and plasma differed significantly. In bone marrow, a macromolecule with molecular mass less than albumin consistently eluted at 13.4 ml (corresponding to 40 kDa), and such a macromolecule was never seen in the plasma samples. In addition, there were even smaller macromolecules in bone marrow fluid that were inconsistently found in plasma constituting a variable but small fraction (<8.3%), of the total area under the elution curve.
|
Induction of a sterile inflammation using Bacto-Tryptone (Fig. 3C) resulted in an increase of macromolecules eluting in the IgM and IgG region. Furthermore, this intervention markedly increased the 13.4-ml peak. In rat bone marrow exposed to repeated cycles of freeze thawing, the major fraction of proteins eluted in volumes larger than that of albumin (12.5 ml), with a major peak at 13.6 ml (Fig. 3D).
Figure 3E shows an elution curve from human plasma, and a representative curve for human bone marrow fluid isolated from a normal subject is shown in Fig. 3F. In the human samples, nearly all macromolecules eluted in the globulin and albumin fractions. The plasma concentrations of albumin and immunoglobulins did not differ between the controls and the patients (data not shown). The most notable difference between human bone marrow fluid and plasma was the smaller albumin-to-globulin ratio for the marrow fluid in normal subjects, averaging 1.50 ± 0.01 (n = 6) compared with the corresponding ratio of 1.81 ± 0.09 (n = 3) obtained in plasma. Similar to the rats, after freeze thawing of human bone marrow, the major part of the macromolecules eluted in volumes higher than that of albumin (Fig. 3G).
Bone marrow fluid contents of hematopoietic cytokines. To determine more precisely whether an increased hematopoietic activity of the bone marrow was reflected in an altered protein composition in the bone marrow fluid, we used Western blotting to detect the two hematopoietic cytokines erythropoietin and G-CSF. It is evident from Figs. 4, A and B, and 5, A and B, that the two cytokines could be identified in the plasma and in the bone marrow fluid in both species. Rat hematopoiesis was then increased either by bleeding the animals or by inducing a sterile peritonitis. Figure 4, C and D, shows that in plasma and bone marrow fluid, the contents of erythropoietin were substantially elevated in the bled rats compared with the controls. Similarly, in inflammatory rats the bone marrow fluid content of G-CSF increased markedly as did the plasma content, albeit to a lesser extent (Fig. 4, E and F).
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Evaluation of the centrifugation method. Bone marrow is a highly specialized cell-rich tissue enclosed in a rigid compartment, and cell compression during centrifugation may lead to extrusion of cellular fluid resulting in a mixture of extracellular and cellular fluid. As in the recent study in chemically induced mammary tumors (22), we decided to use 51Cr-labeled EDTA as a probe in rats to test whether other sources than the bone marrow interstitium contributed to the isolated bone marrow fluid. 51Cr-labeled EDTA (molecular weight 341) is a substance that is not metabolized and is not taken up by cells (13). Addition of cellular fluid to the centrifuged volume should thus be mirrored by a reduced 51Cr-labeled EDTA concentration in the centrifugate relative to plasma. For tissue centrifugation at 
239 g, we found that the ratio of 51Cr-labeled EDTA between centrifugate and plasma was 0.85. By modifications of the centrifugation procedure, we aimed at an extracellular tracer ratio between centrifugate and plasma of 1.0, as found for tumors as well as skin exposed to <424 g. However, even the lowest G force needed to isolate a sample gave a 15% dilution of the isolated fluid, suggesting admixture from another compartment than the extracellular fluid. A further reduction of this ratio when we exposed the tissue to even higher G forces, indicates that an increased dilution occurred, which may be a result of further extrusion of cellular fluid and/or cell damage. Difference in characteristics between the interstitium within the bone marrow and within the tumor and skin may explain why such dilution occurs during centrifugation. As shown here, the amount of extracellular fluid in the bone marrow is 11% of wet weight, which is approximately one-fourth of the corresponding volumes in 9,10-dimethyl-1,2-benzanthracene-induced mammary tumors and skin (22). Another special feature of the bone marrow is the relatively high content of immature cells. Although mature blood cells are able to resist high G forces, e.g., in ordinary separation from plasma, this may not be the case for blood precursor cells. In line with this, isolation of bone marrow extracellular fluid with our method required centrifugation of bone marrow tissue in situ, i.e., within the bone cavity. For obvious reasons we were not able to quantify the contribution to the centrifugate from nonextracellular fluid in humans.
By centrifugation we sampled from the whole extracellular phase, including the vascular volume. Bone marrow plasma volume is low, accounting for <1% of the extracellular fluid volume (5). The fraction of the isolated bone marrow fluid deriving from plasma will accordingly be negligible.
Another potential problem associated with the present method might be cell damage during centrifugation, resulting in contamination of the sample with intracellular proteins. That such contamination occurred to a minor degree is indicated by the peak eluting at 13.4 ml not found in rat plasma. This peak was, however, always <7.2% of the total under the elution curve, suggesting that possible contamination by intracellular proteins was low. To address this question further, we passed through the HPLC column extracts from lysed bone marrow and fluid used for elution of intact marrow. Whereas bone marrow eluate had a HPLC pattern similar to that of plasma except for the peak at 13.4 ml discussed above, marrow lysate had a much higher fraction of low-molecular-mass macromolecules. This is to be expected because a large fraction of the latter consists of intracellular proteins. These experiments, together with the 51Cr-labeled EDTA results, suggest that intracellular fluid without proteins was the fluid that was added to the centrifugate.
Our HPLC data showing that the fraction of large plasma proteins was lower in the bone marrow fluid than in the plasma might indicate that the larger molecular mass immunoglobulins were sieved to some extent during centrifugation. If such sieving occurred, we might have expected an even lower fraction of large molecular mass proteins at increased extracellular matrix compaction likely to occur with increasing G force. Such a change in relative protein content was not observed. Furthermore, in skin, a similar ratio between immunoglobulins and albumin content as that of bone marrow was found in fluid isolated by centrifugation and from wick fluid (22), the latter not likely to be a result of sieving. Taken together, these observations suggest that sieving did not affect the relative distribution of proteins in fluid isolated by centrifugation from bone marrow. This assumption is supported by comparable studies on rat tail tendon where sieving was excluded for various probes spanning a wide range of molecular mass (1).
In rats as well as in the human subjects, the colloid osmotic pressures in plasma were similar to those found in previous studies (17, 22). COP relative to plasma, however, was significantly higher in bone marrow fluid than observed for subcutaneous extracellular fluid (22) but comparable to the corresponding relative pressure in tumors (21, 22), a tissue with high capillary permeability and dysfunctional lymphatics (10, 18), i.e., similar to the bone marrow with its leaky sinusoides and lack of lymphatic vessels.
Local levels of cytokines reflect hematopoietic challenges. To further evaluate centrifugation as a valid method to isolate bone marrow extracellular fluid, we specifically examined its content of erythropoietin and G-CSF. The levels of these two hematopoietic cytokines vary according to the prevailing hematopoietic activity with high levels seen during augmented erythro- or leukopoiesis. Our present data show that erythropoietin was markedly increased in bone marrow extracellular fluid in hemorrhagic rats. Inflammation, induced as aseptic peritonitis, led to a substantial increase of G-CSF in the bone marrow extracellular fluid. The cytokine levels were apparently larger in bone marrow extracellular fluid than in plasma. If anything, the cytokine contents of the extracellular fluid may represent an underestimate due to a possible dilution of the samples, as discussed above. This assumption is supported by the increase in the peak eluting at 13.4 ml during sterile inflammation (Fig. 3C), a finding suggesting a higher proportion of cell fluid in the centrifugate during this condition. Although being outside the scope of the present study, a further characterization of the protein(s) eluting in this peak may be of interest.
Functional aspects of bone marrow extracellular fluid composition. Being situated between the vascular compartment and the bone marrow cells, the composition of the bone marrow interstitium probably affects locally both the structure and the transport of nutrients, growth factors, and signaling molecules essential for an adequate hematopoietic activity. For example, the hematopoietic cytokines erythropoietin and G-CSF modulate bone marrow vascular resistance, at least partly, by locally releasing nitric oxide (4, 8). Furthermore, data indicate a tight regulation of bone marrow interstitial pressure (5). Moreover, systemic hemodynamic disorders such as cardiac failure, significantly attenuated murine bone marrow hematopoiesis via upregulation of local regulators of hematopoiesis such as tumor necrosis factor-
and the apoptosis effector molecule Fas (9). An increasing number of cytokines have been identified with the ability to modify blood cell egress (12). Recent experiments indicate that some of these cytokines also regulate neovascularization, in particular during development and progression of hematopoietic malignancies (6, 19). Collectively, these data emphasize the need for refinement of methods to obtain such regulatory substances at the local level; i.e., in the bone marrow interstitium.
Blood cancers, such as the acute leukemias and multiple myelomas, have poor prognosis, even with intensive chemotherapy followed by stem cell transplantation. In addition to the need for more mechanism-based therapeutic regimens, the early determination of minimal residual disease in these patients requires improved detection of marker molecules within the bone marrow microenvironment (3). Such detection will ultimately depend on a detailed knowledge of the distribution and concentration of putative hematopoietic regulators in this compartment. Consequently, the current centrifugation method offers a novel approach to such analyses.
|
ACKNOWLEDGMENTS |
|---|
GRANTS
The study was financed in part by grants provided by the Locus on Cardiovascular Research at University of Bergen, The Research Council of Norway, The Norwegian Research Council on Cardiovascular Diseases, The Norwegian Cancer Society, and the Throne-Holst Foundation.
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  E. Semaeva, O. Tenstad, A. Bletsa, E.-A. B. Gjerde, and H. Wiig Isolation of rat trachea interstitial fluid and demonstration of local cytokine production in lipopolysaccharide-induced systemic inflammation J Appl Physiol, March 1, 2008; 104(3): 809 - 820. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. O. Iversen and H. Wiig Tumor Necrosis Factor {alpha} and Adiponectin in Bone Marrow Interstitial Fluid from Patients with Acute Myeloid Leukemia Inhibit Normal Hematopoiesis Clin. Cancer Res., October 1, 2005; 11(19): 6793 - 6799. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. E. Markhus and H. Wiig Isolation of interstitial fluid from skeletal muscle and subcutis in mice using a wick method Am J Physiol Heart Circ Physiol, November 1, 2004; 287(5): H2085 - H2090. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
