Relationship of bursal anti-steroidogenic peptide (BASP) and histone H1

Life Sciences 71 (2002) 3071 – 3079 www.elsevier.com/locate/lifescie

Relationship of bursal anti-steroidogenic peptide (BASP) and histone H1 Gary Garcı´a-Espinosa a, Randy W. Moore b, Luc R. Berghman c, Billy M. Hargis a,* a

Department of Poultry Science, University of Arkansas, Fayetteville, AR 72701, USA b USDA-ARS, 2881 F&B Road, College Station, TX 77845, USA c Department of Poultry Science, Texas A&M University, College Station, TX 77843, USA Received 1 March 2002; accepted 2 July 2002

Abstract Previous in vitro research from our laboratory has demonstrated the existence of a protein purified from the chicken bursa of Fabricius, with potent antisteroidogenic and antiproliferative action on granulose cells and lymphocytes, respectively called Bursal anti-steroidogenic peptide (BASP). This protein is heat-labile, basic, and amino- and carboxy-terminus blocked. In highly purified form, the protein presents as a doublet on SDS –PAGE electrophoresis with an apparent MW of f 29 and f 32 kDa. Recently, Nanoflow Q – TOF Mass Spectrometry amino acid sequencing allowed determination of a convincing partial amino acid sequence, strongly suggesting a probable relationship of BASP with histone H1. Bursal cDNA expression library screening, using an antibody produced against BASP, also identified a clone with a sequence matching histone H1. Presently, we have demonstrated that SDS – PAGE electrophoresis of highly purified and bioactive BASP, and commerciallyavailable calf thymus derived histone H1, produced similar doublets at approximately the same apparent MW, and that the electrophoretic profile of these 2 preparations were strikingly similar following 2 dimensional gel electrophoresis. The BASP doublet produced on SDS – PAGE was recognized by a commercially available monoclonal antibody recognizing a highly conserved region of histone H1. Furthermore, calf thymus histone H1 was found to suppress mitogen-stimulated chicken B-cell proliferation in a concentration-related manner, similar to the action of BASP. These data indicate that BASP shares substantial structural homology with, and may be identical to, histone H1. D 2002 Elsevier Science Inc. All rights reserved. Keywords: Histone H1; Bursa of fabricius; Anti-steroidogenic; Anti-DNA synthesis; B-lymphocytes; Chicken

* Corresponding author. Tel.: +1-479-575-1527; fax: +1-479-575-8490. E-mail address: [email protected] (B.M. Hargis). 0024-3205/02/$ - see front matter D 2002 Elsevier Science Inc. All rights reserved. PII: S 0 0 2 4 - 3 2 0 5 ( 0 2 ) 0 2 1 7 0 - 7

3072

G. Garcı´a-Espinosa et al. / Life Sciences 71 (2002) 3071–3079

Introduction The bursa of Fabricius (BF) is a discrete lymphoid organ, unique to birds, and is the primary site of B-lymphocyte ontogeny and early proliferation [1–3]. The existence of this unique anatomical site of B-lymphocyte development in chickens provided the opportunity for the first evidence of the existence of the humoral immune system, distinct from cell-mediated immunity [1,2,4,5,6]. Development of the BF begins approximately at day 8 of embryonic life [7] and continues through 4 to 11 weeks after hatch in chickens [8]. The expansion, differentiation, and emigration of B-lymphocytes to peripherial lymphoid organs occurs prior to regression of the BF and onset of sexual maturity in birds [8]. In addition to this well-demonstrated role of the BF in establishment of humoral immunity, several lines of evidence have prompted investigators to propose a possible endocrine role for the BF. The temporal association of bursal regression with pubertal development, coupled with evidence that sex or corticosteriods can cause premature bursal involution prompted early speculation that the BF was at least partially controlled by the endocrine system [9,1011]. Reports that surgical ablation of the BF could accelerate the onset of sexual development [12,13,14,15] further strengthened speculations of an endocrine function of the BF. The observation of discrete cells near the corticomedullary border of the BF with a secretory appearance further suggested a possible endocrine role [16,17]. Postulating a possible negative feedback involvement with steroid hormone-producing gonadal and perhaps adrenal cortical cells, Pedernera and colleagues demonstrated that crude aqueous BF extracts could suppress testosterone release from rat Leydig cells [18]. More recently, our laboratories demonstrated that a single highly purified heat-labile protein from chicken BF extracts, named Bursal Anti-steroidogenic Peptide [19] was capable of potent in vitro anti-steroidogenic activity in chicken ovarian granulosa cell cultures and on adrenocortical cells isolated from the chicken, dog, and pig [20]. This molecule was also shown to negatively affect mitogen-induced DNA synthesis in isolated lymphocytes from chicken, rat, and human sources [21,22] and to increase growth hormone and prolactin from isolated chicken anterior pituitary cells in vitro [23]. Previous micro-sequencing attempts have been frustrated because of dual amino- and carboxy–terminal blockage [19]. Very recently, we obtained 2 short (6 and 8 aa) partial sequences from individual digests of the upper (apparent 34 kDa) and lower (apparent 29 kDa) PAGE bands, which consistently have been observed with bioactive highly purified BASP [22], which were consistent with known sequences of histone H1. Presently we have evaluated the possible physicochemical and biological relationship of purified bioactive BASP to commercially available calf thymus histone H1.

Methods Extraction and purification of BASP A modification of previously-published procedures for BASP extraction and purification were used in the present studies [21,22]. Briefly, bursae were collected from 7-week-old chickens immediately after death and held at 76 jC prior to extraction. Tissue (200 g) was homogenized with 2 parts 15% trifluroacetic acid (TFA). Homogenized material was centrifuge twice at 37 000  g for 20 minutes at 4 jC. The resulting supernatant was then carefully removed, avoiding contamination with surface lipid,

G. Garcı´a-Espinosa et al. / Life Sciences 71 (2002) 3071–3079

3073

and loaded onto 5 mM TFA-equilibrated 2.5  10 cm preparative C-18 Mega bond elut columns (Varian Catalog 1225–6031). Cartridges were stepwise eluted (15 ml) with 0, 5, 30, 60, or 80% acetonitrile (ACN) with 5 mM TFA and eluent was collected in polypropylene tubes, and dried by vacuum centrifugation. The 30% ACN fraction, previously shown to contain most of the bioactive BASP [21,22] was evaluated by SDS–PAGE, and the expected protein doublets near 30 kDa were observed . This fraction was then subjected to cation exchange chromatography. Briefly, dried material was hydrated in double distilled water before passing through a 1.5  20 cm glass column containing twenty–five grams of carboxy–methyl accell plus CM cation exchange medium 300 A (Waters corporation, Catalogue 10740) equilibrated with 3 M NH4HCO3 and washed with 10 times volume of double distilled water. The column was then eluted stepwise with 10 times volume of increasing concentrations of NH4HCO3. The 0.75 M NH4HCO3 eluate was retained and concentrated in a YM-30 kDa centriplus unit adding double distilled water to the sample two times and centrifuged at 2,000  g for 45 minutes (Amicon, Catalogue 4412), dried by vacuum centrifugation and stored at 20 jC until use. The presence of the expected protein doublet near 30 kDa was again observed by SDS–PAGE and bioactivity was confirmed as described below. SDS–PAGE and western blot Samples of biologically active BASP or calf thymus histone H1 (Cat. 382150, Calbiochem) were mixed in sample buffer (62.5 mM Tris HCl, pH 6.8, 20% glycerol, 2% SDS), boiled for 5 minutes, and loaded on a 12.5% acrylamide/Bis gel (Laemmli’s method). Following electrophoresis, (Protean II xi Cell system, BIO–RAD, Catalog 165–1934) resulting gels were either stained with coomassie blue G250 or transfered to nitrocellulose (Trans-Blot, BIO–RAD Cat. 170–3939). Nitrocellulose membranes were soaked in blocking solution (10 mM Sodium Phosphate, pH 7.2, 1 mM NaCl, 0.1% Tween-20, 5% non-fat dry milk) for at least 1 hour at room temperature prior to immunostaining. Antibodies were diluted in this blocking solution. A monoclonal antibody clone AE-4 (ICN Biomedicals, Inc. Cat. 69331, antibody subtype IgG2a) that has previously been shown to specifically recognize histone H1 of several animal and plant species was used at a 3000  dilution with a 2 hr incubation at room temperature. A similarly diluted irrelevant mouse monoclonal antibody anti-proliferating cell nuclear antigen, clone PC10 (Sigma, Cat. P8825) of the IgG2a subtype was used as a control for possible nonspecific attachment. Mouse antibody binding was detected using a commercial secondary antibody kit (Vectastin Universal Elite ABC Kit Cat.PK-6200) according to the manufacturer’s instructions with development of the peroxidase reaction with diaminobenzidine (DAB; 0.1 Ag/ml PBS, pH 7.2, 0.012% H2O2). 2-dimensional PAGE Samples of calf thymus histone H1 and BASP were subjected to 2 Dimensional electrophoresis [24] using a X Cell Mini-gel Electrophoresis system (Invitrogen). 10 Ag or each sample was dissolved in cathode buffer sample (Invitrogen, Cat.LC5311) with 2% h-mercaptoethanol and boiled for 5 minutes before loaded on a pre-cast pH 3–10 isoelectric focusing gel (IEF) (Invitrogen, Cat.EC6655A2). The gel was then cut and embedded into a pre-cast 4–20% Tris–glycine Zoom gel (Invitrogen, Cat.EC60261) Finished gels were stained and developed using a commercially-available colloidal blue staining kit following manufacture’s instructions (Invitrogen, Cat.LC6025).

3074

G. Garcı´a-Espinosa et al. / Life Sciences 71 (2002) 3071–3079

Bioassay The bioassay, based on suppression of mitogen-induced DNA synthesis in neonatal chicken Blymphocytes, has been previously reported [21,22]. Approximately 20 one-day-old single-comb white Leghorn (SCWL; Hy-Line W36R) chicks were obtained from a local commercial hatchery (Hy-Line International, Bryan, TX). Immediately following CO2 asphyxiation, the feathers were moistened with 70% ethanol, and chicks were placed under a laminar flow hood for dissection and excision of cloacal bursae. The bursa of Fabricius of each chick was excised at the proctodeal junction and immediately placed into a sterile 50 mL conical centrifuge tube containing 25 mL of RPMI 1640 medium [ with 0.1% bovine serum albumin (Sigma, Cat. A9418), 100 units ml 1 penicillin and 100 Ag ml 1 streptomycin (Gibco, Cat. 15070–014), and 50 Ag ml 1 gentamicin]. All subsequent steps in the bursal lymphocyte isolation procedure were performed on wet ice at approximately 4 jC. After bursae were removed from all carcasses, they were transferred to a sterile petri dish containing approximately 5 ml of the complete RPMI 1640 medium. The bursae of Fabricius were gently teased apart with 18 gauge needles until a suspension of small tissue fragments was obtained. Tissue fragments were then transferred to a sterile 50 ml conical tube where they were gently triturated with four sterile 5 cc syringes with progressively smaller inlets (approximately 8, 6, 4, or 2 mm inlet sizes produced by trimming the luer portion of the syringe with a scalpel). Beginning with the syringe with the largest inlet, tissue was triturated until the resulting cell suspension was noticeably opaque, at which time, tissue fragments received a total of about 30 triturations per syringe. The resulting cell suspension was filtered to remove clumps of cells or tissue fragments by covering a sterile 50 ml conical tube with nylon mesh of 60 Am pore size (Nytex,Tetko Inc.). The filtered cell suspension was then pelleted by centrifugation for 10 min at 350  g at 4 jC. Cells were washed by resuspension in fresh complete RPMI 1640 medium (4 jC) and repelleted by centrifugation. Cells were washed a total of three times prior to assessment of number and viability. Immediately following the final washing step, a 100 Al aliquot of the cell suspension was subjected to the live–dead staining procedure of Freshney [25], and the final concentration of cells in suspension was adjusted to 4  106 viable cells/ml with fresh RPMI. Viability of isolated cells was > 90% in all experiments. Cell culture was carried out in 96-wel polystyrene tissue culture plates (Falcon). Treatments, mitogen, cells, and thymidine methyl3H (ICN Pharmaceuticals; 1:199 dilution of thymidine methyl3H, specific activity 6.7 Ci/mmol and 1 mCi/ml, with cell culture medium) was added to each well in 50 Al volumes so that the final volume in all wells in each experiment was 250 Al. Six replicate wells were used for each treatment. Cultures were incubated at 37 jC, with saturated humidity and 5% CO2, for 24 hr. 2  105 Isolated B-lymphocytes per well were exposed to 200 ng/ml of mitogen (phorbol 12,13-dibutyrate; Sigma) and/or serial double dilutions of 5 Ag of either BASP or calf thymus histone H1 for 16 hr prior to the addition of thymidine methyl3H. Following these initial incubation periods, cells were pulsed with 1 ACi/ml of thymidine methyl3H for the final 8 hr of the entire culture period. At termination, cells were harvested onto filter mats (Skatron Instruments) using a semi-automatic 12 well cell harvester (Skatron Instruments). Filter mats were dried, the filter discs corresponding to each well were placed into 7 mL polypropylene scintillation vials, and 5 mL of an aqueous-based scintillation counting cocktail (Ultima Gold, Packard Instrument) was added. Radioactivity was detected for 2 min in a standard liquid scintillation counter (Beckman Instruments). Counts were automatically quench corrected and data were reported as disintegrations per minute (DPM).

G. Garcı´a-Espinosa et al. / Life Sciences 71 (2002) 3071–3079

3075

Results and discussion In early attempts at purification and characterization of BASP, we reported that some antisteroidogenic bioactivity was observed in ultrafiltration fractions less than 5 kDa [19]. However, subsequent studies clearly demonstrated that the majority of BASP bioactivity was retained by 30 kDa membranes during dialysis and that bioactive preparations were consistently associated with doublet protein staining on SDS–PAGE gels at apparent average MW of just over 30 kDa [22]. While it is possible that more than one molecule with the described bioactivities existed in bursal extracts, there are other potential explanations for these differing reports. Later extraction methods used a more selective solvent (described above), which might not have recovered the bioactivity associated with the milder extraction process used in the earlier studies [19,20]. Alternatively, it is possible that the relatively lengthy ultrafiltration process used in the early report could have allowed for enzymatic cleavage resulting in bioactive fragment(s) of a larger parent molecule. As the early description of BASP size was based solely on ultrafiltration characteristics, it is also possible that membrane integrity was compromised allowing leakage of larger MW proteins. Regardless, it is now clear that bioactivity is associated with the larger MW estimates of roughly 30 kDA, and that protein doublets described by Caldwell et al. (1999) [22] consistently and uniquely are observed in highly purified BASP preparations. Recently, Nanoflow Q–TOF Mass Spectrometry amino acid sequencing allowed determination of a short but convincing partial amino acid sequence, suggesting a probable relationship of BASP with histone H1 (data not shown). Bursal cDNA expression library screening, using an antibody produced against BASP, also identified a clone with a genetic sequence consistent with histone H1 (unpublished). For these reasons we have presently explored the possibility of a relationship between BASP and histone H1. Previous reports have indicated that histone H1 appears as a doublet on SDS–PAGE with an average apparent MW of 33 kDA [26], similar to our observations with purified BASP preparations [22]. A similar pattern of protein doublets was observed following staining of simultaneously electrophoresed purified BASP and commercially available calf thymus histone H1 (Fig. 1). The appearance of a doublet with histone H1 is likely due to post-translational modifications [27]. Differences in the relative staining intensity of the two bands in each of the compared preparations could be due to relative differences in tissue post-translational activity between the calf thymus and the chicken bursa of Fabricius. A commercially available monoclonal antibody, which specifically recognizes a conserved histone H1 epitope across a wide range of plant and animal species, was then used to further examine the apparent similarities between BASP and calf thymus histone H1. Western Blot analysis of a SDS PAGE gel similar to that shown in Fig. 1 indicated that the each of the double bands present in BASP and histone H1 were recognized by the same monoclonal antibody indicating that these preparations share this epitope (Fig. 2). To evaluate the possibility that other proteins possibly responsible for BASP bioactivity were coeluting with histone H1 on SDS–PAGE, we compared the 2-dimensional electrophoretic pattern of BASP and calf thymus histone H1 (Fig. 3). With either preparation, major proteins were identified at f 31 kDa at f pH 6, with staining also at f 21.5 kDa were apparent at either f pH 8 or 4. The striking 2Dimensional electrophoretic pattern similarities between these 2 preparations is also consistent with the postulate that BASP is at least closely related to histone H1. No evidence of any major hidden protein in the BASP preparation, not found in the commercial calf thymus histone H1, was detected.

3076

G. Garcı´a-Espinosa et al. / Life Sciences 71 (2002) 3071–3079

Fig. 1. SDS – PAGE comparison of histone H1 (lane 2) and BASP (lane 3). Each sample represents 2 Ag and pre-stained standards are included (lane 1). Gels were stained using Coomassie blue.

Because of the similarities between histone H1 and BASP, and because histone H1 is highly conserved and ubiquitous in animal tissues in association with its known function in DNA supercoiling [28], we simultaneously evaluated the ability of commercially available calf thymus histone H1 and BASP to suppress mitogen-stimulated neonatal chicken bursal lymphocyte DNA synthesis (Fig. 4). Both preparations suppressed mitogen-stimulated bursal lymphocyte DNA synthesis in a concentrationdependent fashion. Although both preparations were highly efficacious in this bioassay, the calf thymus histone H1 was apparently less potent in two independent cell cultures (Fig. 4, data for replicate culture not shown). Possible explanations of apparent potency differences include differences during isolation and/or storage, impurities (e.g. salt content) in the preparations, or sequence heterology in the effector region of the molecule. In support of this latter possibility, the amino acid sequence of bovine histone H1 is f 61 to 67% homologous with chicken histone H1, and the most conserved region of the molecule, the globular domain, is approximately 87% homologous between these species. Differences in potency could also be related to differing degrees of post-translational modification in the source tissues. For

Fig. 2. Recognition of calf thymus histone H1 (2) or BASP (3) by the mouse monoclonal antibody anti-histone H1 clone AE-4. Irrelevant IgG2a (the same subclass) was used as negative control for each of the respective preparations (4,5). Following SDS – PAGE (5 Ag each preparation) proteins were transfered to a nitrocellulose membrane, which was divided for immunostaining and re-assembled for photography (between lanes 3 and 4).

G. Garcı´a-Espinosa et al. / Life Sciences 71 (2002) 3071–3079

3077

Fig. 3. 2-Dimensional gel electrophoresis comparing calf thymus histone H1 (Left panel) and BASP (Right panel). 10 Ag per sample was loaded in an IEF gel and embedded in a Tris – Glycine gel. Proteins were stained by colloidal blue.

example, differing degrees of glycosolation could in fact alter relative potency. Although not directly examined in the present experiments, the relative intensity of upper and lower bands of the doublets observed in Fig. 1 may suggest some post-translational processing differences between these different tissues and different species. Taken together, these results strongly suggest that BASP is similar to histone H1. During the last few years, several in vitro hormone-like activities of histones or histone fragments have been described [29]. It is also interesting to note that several of the purification protocols that have evolved for histone H1 are similar to those that have independently been developed for BASP [30]. The best-documented role for

Fig. 4. Effect of increasing concentrations of BASP (Black bar) or calf thymus histone H1 (ctH1) (White bar) on phorbolstimulated DNA synthesis in bursal cells obtained and pooled from 20 one-day-old chickens. Basal represents DNA synthesis in the absence of the phorbol mitogen and PDB represents effects of the mitogen without either of the evaluated preparations. Bars represent mean DPM F SE. The results represent one of two independent experiments with similar results.

3078

G. Garcı´a-Espinosa et al. / Life Sciences 71 (2002) 3071–3079

histones is related to nuclear DNA organization and gene regulation [28]. Although histones have been recently detected in the circulation of mammals during some pathological events [31,32], the possible role of these molecule(s) as signal proteins remains to be elucidated.

Acknowledgements Thanks to Dr. Narayan C. Rath, USDA/ARS, Poultry Science Department, University of Arkansas for his assistance with 2-dimensional electrophoresis.

References [1] Glick B, Chang TS, Jaap RG. The bursa of Fabricius and antibody production. Poultry Science 1956;35:224 – 6. [2] Cooper MD, Petterson RDA, South MA, Good RA. The functions of the thymus system and the bursa system in the chicken. Journal of Experimental Medicine 1966;123(1):75 – 102. [3] Lydyard P, Grossi CE, Copper MD. Ontogeny of B cell in the chicken: Sequential development of clonal diversity in the bursa. Journal of Experimental Medicine 1976;144(1):79 – 97. [4] Chang TS, Rheins MS, Winter AR. The significance of the bursa of Fabricius in antibody production in chickens. Poultry Science 1957;36:735 – 8. [5] Mueller AP, Wolfe HR, Meyer RK. Precipitin production in chickens. XXI. Antibody production in bursectomized chickens and in chickens injected with 19-nortestosterone on the fifth day of incubation. Journal of Immunology 1962;85:172 – 9. [6] Mueller AP, Wolfe HR, Meyer RK, Aspinall RL. Further studies on the role of the bursa of Fabricius in antibody production. Journal of Immunology 1962;88:354 – 60. [7] Ratcliffe MJH, Lassila O, Reynolds J, Richard J, Pink L, Vainio O. A re-evaluation of the function of the bursa of Fabricius. Progress in Clinical and Biological Research 1987;238:3 – 14. [8] Glick B. Normal growth of the bursa of Fabricius in chickens. Poultry Science 1956;35:843 – 51. [9] Kirkpatrick CM, Andrews FN. The influence of sex hormones on the bursa of Fabricius in the ring necked pheasant. Endocrinology 1944;34:340 – 5. [10] Glick B. Experimental modification of the growth of the bursa. Fabricius Poultry Science 1957;36:18 – 23. [11] Meyer RK, Rao MA, Aspinall RL. Inhibition of the development of the bursa of Fabricius in the embryos of the common fowl by 19nortestosterone. Endocrinology 1959;64:890 – 7. [12] Jolly J, Pezard A. La castration retarde involution de la bourse de Fabricius. Comptes Rendus des Se´ances de la Socie´te´ de Biologe et de Ses Filiales 1928;98:379. [13] Riddle O. Growth of the gonads and bursa of Fabricius in doves and pigeons with the data for body growth and age at maturity. American Journal of Physiology 1928;86:243 – 65. [14] Pedernera EA, Romano M, Besedovsky HO, Aguilar MDC. The bursa of Fabricius is required for hormonal development in chickens. General and Comparative Endocrinology 1980;42(3):413 – 9. [15] Mashaly MM. Bursectomy and its influence on circulating corticosterone, triiodothyronine and thyroxine in immature male chickens. Poultry Science 1984;63(4):798 – 800. [16] Olah I, Glick B. Secretory cell in the medulla of bursa of Fabricius. Experientia 1978;34(12):1642 – 3. [17] Glick B, Olah I. Contribution of a specialized dendritic cell, the secretory cell to the microenviroment of the bursa of Faricius. Prog Clinical Biology Research 1987;238:53 – 66. [18] Pedernera E, Aguilar MC, Romano M. Bursa of Fabricius inhibits in vitro chorionic gonadotrophin response of the chick testis. General and Comparative Endocrinology 1985;57(1):124 – 9. [19] Byrd JA, Hayes TK, Wright MS, Dean CE, Hargis BM. Detection and partial characterization of an anti-steroidogenic peptide from the humoral immune system of the chicken. Life Sciences 1993;52(14):1195 – 207. [20] Byrd JA, Dean CE, Hargis BM. The effect of the humoral immune system derived bursal anti-steroidogenic peptide

G. Garcı´a-Espinosa et al. / Life Sciences 71 (2002) 3071–3079

[21]

[22]

[23]

[24] [25] [26] [27] [28] [29] [30]

[31] [32]

3079

(BASP) on corticosteroid biosynthesis in avian, porcine, and canine adrenal cortical cells. Comparative Biochemical and Physiology 1994;108(2):221 – 7. Caldwell DJ, Caldwell DY, McElroy AP, Manning JG, Hargis BM. Bursal Anti Steroidogenic Peptide Suppression of Mitogenesis in Chicken, Rat, and Human Peripheral Blood Lymphocytes. Developmental and Comparative Immunology 1998;22(5 – 6):613 – 20. Caldwell DJ, Dean CE, McElroy AP, Caldwell DY, Hargis BM. Bursal Anti-Steroidogenic Peptide (BASP): Modulation of Mitogen-Stimulated Bursal -Lymphocyte DNA Synthesis. Comparative Biochemistry and Physiology 1999;123(4): 385 – 91. Caldwell DJ, Manning JG, Caldwell DY, McElroy AP, Hargis BM, Porter TE. Effect of bursal antisteroidogenic peptide (BASP) on chicken embryonic pituitary secretion of growth hormone (GH) and prolactin (PRL): evaluation in a reverse hemolytic plaque assay (RHPA). Domestic Animal Endocrinology 1999;16(1):81 – 7. Dunbar BS, Kimura H, Timmons TM. Protein analysis using high-resolution two-dimentional polyacrylamide gel electrophoresis. In: Deutscher MP, editor. Guide to protein purification. San Diego: Academic press; 1990. p. 441 – 59. Freshney RI. Quantitation and experimental design. In: Freshney RI, editor. Culture of Animal Cells. New York: A.R. Liss; 1983. p. 207 – 9. Grunwald D, Lawrence JJ, Khochbin S. Accumulation of histone H1j during early Xenopus laevis development. Experimental Cell Research 1995;218(2):586 – 95. Berger SL. An embarrassment of niches: the many covalent modifications of histones in transcriptional regulation. Oncogene 2001;20(24):3007 – 13. Ramakrishnan V. Histone structure and the organization of the nucleosome. Annual Review in Biophysics and Biomolecular Structure 1997;26:83 – 112. Brown OA, Sosa YE, Goya RG. Histones as extracellular messengers: Effects on growth hormone secretion. Cell Biology International 1997;21(12):787 – 92. Yasutmoto Y, Suga T, Wada S, Kosano H, Takishima K, Mamiya G, Kugai N, Nagata N. Purification and partial sequencing of inhibitory factor on renal membrane adenylate cyclase in pancreatic cancer extract: identity with histones H1B or H1D. Biochemical and Biophysical Research communications 1991;176(1):255 – 61. Rumore PM, Steinman CR. Endogenous circulating DNA in systemic lupus erythematosus. Occurrence as multimeric complexes bound to histone. Jounal of Clinical Investigation 1990;86(1):69 – 74. Bab I, Gazit D, Chorev M, Muhlrad A, Shteyer A, Greenberg Z, Namdar M, Kahn A. Histone H4-related osteogenic growth peptide (OGP): a novel circulating stimulator of osteoblastic activity. EMBO 1992;11(5):1867 – 73.