Epitope recognition in histone H1 by SLE autoantibodies in the presence of a DNA-ligand.

Auroimmuniry. 1992, Vol. 12, pp. 167-174 Reprints available directly from the publisher Photocopying permitted by license only

0 1992 Harwood Academic Publishers GmbH Printed in the United Kingdom

EPITOPE RECOGNITION IN HISTONE H1 BY SLE AUTOANTIBODIES IN THE PRESENCE OF A DNA-LIGAND PAULINE CREEMERS', MARC MONESTIER2 and LOTHAR BOHM' 'Department of Radiotherapy. University of Stellenhosch, Faculty of Medicine, P.O. Box 19063, Tygerberg 7505, Republic of South Africa; 'Garden State Cancer Center and Center for Molecular Medicine and Immunology, I Bruce Street, Newark, NJ 0 7 / 0 3 , U.S.A.

Autoimmunity Downloaded from informahealthcare.com by Chulalongkorn University on 01/08/15 For personal use only.

(Received September 9,1991; in final form January 23,1992)

To investigate the specificity of anti HI antibodies peptides from the N- and C-domain of HI and the synthetic oligonucleotide (AT), were complexed. Circular dichroism (CD) spectroscopy indicated that the free peptides HI(I-I6), Hl(204-218) and C(121-210) in low salt buffer assume a random structure but become helical when bound to the oligonucleotide. The structured and unstructured HI fragments were then analyzed by enzyme linked immunosorbent assay (ELISA) with anti-HI antibodies in sera from patients with systemic lupus erythematosis (SLE) and with the monoclonal anti-HI antibody MRA-I2 derived from MLR Ipr/lpr autoimmune mice. Binding of these antibodies to Hl(204-218) and C was inhibited to a level of 50% when these HI peptides were complexed with (AT)6. When the same antibody was tested with HI fragment GC(34-2 lo), attachment to oligonucleotide (AT)6 did not influence antibody binding. Competition studies with liquid phase GC and C antigen against solid phase GC and C indicated that liquid phase GC was more efficient in displacing antibody binding reactivity than liquid phase C. The displacement effect of both liquid phase antigens was greatest against solid phase C. We conclude that anti-HI autoantibodies are directed against an epitope located near the junction of the G- and C-domain which is exposed and not masked when HI is bound to DNA. KEY WORDS: Anti histone antibodies, anti nuclear antibodies, systemic lupus erythematosus (SLE), histone HI.

The antibody stimulating structure in SLE could be free H1, or H I bound to DNA as chromatin. Free HI has been found in low quantities in serum' and autoantibodies that react with core histone can also recognize an unknown cell membrane antigen'. The structure of free H1 in solution and of H 1 bound to DNA is different: In the absence of DNA the important C-terminal tail of H1 behaves like a freely mobile random coil"' but this region shows great helix potential"-'3 which can be demonstrated by helix promoting agents and by charge n e u t r a l i ~ a t i o n ' ~ . ' ~ . We performed circular dichroism (CD) experiments of complexes between HI peptides and oligonucleotide (AT),. The results suggest that.the C-terminal tail assumes helicity upon binding to linker DNA analogue (AT),. We therefore examined the reactivity of H 1 fragments with autoantibodies before and after oligo (AT), binding. The results presented below give insight into the fine specificity of antibodies and the nature of the stimulating antigen.

INTRODUCTION Histone H1 and histone H2B are the most prominent antigens of chromatin. These determinants are expressed as autoantibodies with a vafiety of other nuclear antigens in the serum of patients with systemic lupus erythematosus (SLE)132. In the understanding of the disease the identification of the antibody stimulating structures is of great importance. In the search for the stimulating antigen, sera have been tested with a variety of H1 peptides. ELISA and immunoblotting tests revealed that virtually all H 1 reactive sera reacted to the large C-terminal fragment. Reaction of sera with the amino terminal part of the molecule N was consistently weak and no reactivity was found to the central globular G domain',"'. The mouse monoclonat antibody MRA- 12, derived from MLR Ipr/lpr mice which develop a lupus-like syndrome reacted strongly in the ELISA with GC, a peptide comprising the G- and C-domain at residues 34-210, and to a lesser extent to the C terminal end. However, in the Western blot the MRA-12 antibody reacted only to the GC peptide, indicating that the antigenic determinant is fully expressed only in GC'. These findings suggest the involvement of a large number of residues in the H1 epitope.

MATERIALS AND METHODS Pa tien ts

Sera were derived from patients diagnosed with SLE according to the 1982 revised criteria for the classifi-

Cvrrespondence: Dr. L. Bohm.

I67

'

P. CREEMERS, M. MONESTIER A N D L. BOHM

I 68

Hl(204-218), 0.78 pglml; and (AT),, 1.05 pg/ml. When peptides combined with (ATk were tested, the above peptide concentrations were used, combined with (AT), in a molar ratio of 1 peptide to 2 (AT), for C, GC and NG, and 1 to 1 for Hl(1-16) and Hl(204-218). Fifty pl of antigen was added to each well in a polystyrene microhemagglutination plate Preparation of purified H I , H I peptides and (AT)h (Cooke microtiter 129 A, Dynatech Laboratories) and incubated overnight at 4°C. Thereafter the test was Purified H1 was prepared from calf thymus as described". An overview of the subunits of H1 is performed in phosphate-buffered saline (PBS-)Tween given in Table 1. Peptide G was prepared by trypsin (0.05%): After washing, patient sera or monoclonal cleavage"; NG and C peptides were prepared by antibody were added (50 pullwell) and the plates were thrombin cleavage'', and GC was prepared by cleav- incubated for 1 hr at 37°C. After washing, the plates age with protease from mouse submaxillary gland2'. were incubated with 100 pl/well biotinylated goat These peptides refer to the H1 sequence of the calf anti-human Ig (H&L) or biotinylated rabbit antithymus histone HI subfraction CTL-I which com- mouse Ig (H&L) (Zymed Laboratories Inc.). After an prises 210 residues the full sequence of which has additional incubation of 1 hr at 37"C, plates were been giveni5. Peptides H 1( 1-1 6) and H l(204-2 18) washed and 100 pullwell alkaline-phosphatase streptawere obtained by peptide synthesis with the solid vidin (Zymed Laboratories, Inc.) were added. The phase method of Merrifield as described2' and refer to plates were incubated for another 30 min at 37°C. After washing 100 pl/well p-nitrophenyl disodium the human spleen histone variant H 1b2*. The oligonucleotide 5'-(AT),-3' ( (AT), 1 was syn- phosphate (1 mg/ml, Sigma) substrate in 1 M diethathesized in an Autogen 6500 synthesizer using phos- nolamine, 0.5 mM MgC& was added; the reaction was stopped with 40 pl/well 3 N NaOH. Optical density phoimidite methodology. (O.D.) was determined at 405 nm in an EIA reader model EL-307 (Bio-Tek Instruments, Inc.). Circular dichroism (CD)


cation of SLEi6.Sera which were reactive to purified H1 and which showed little or no reactivity to ss and ds DNA as determined by ELISA were selected. Samples were obtained from the Microbiology Laboratory, Tygerberg Hospital, Tygerberg, South Africa.

Measurements were performed in low salt buffer, in which H1 peptides assume a random coill4. Low salt buffer consisted of 1 mM Na phosphate, 0.2 mM Na2EDTA, pH 7.4. CD spectra were determined using a Jasco J-40A automatic recording spectropolarimeter at a time constant of 64sec integration time and a wavelength expansion of 5 nm per cm. Spectra were recorded at wavelengths of 340 to 190 nm.

Enzyme-linked immunosorhent assay (ELISA) This method was performed as described23. Histones or their fragments were diluted in carbonate-bicarbonate coating buffer (pH 9.6) at equimolar concentrationq G and C, 5 pg/ml; GC, 10.5 pg/ml; NG, 6.4 p g/ml; H1, 14pg/ml; HI( 1-16), 0.89 pg/ml;

Testing of sera In total, 34 sera from SLE patients and 10 sera from normal controls were tested. Prior to testing, sera were absorbed with ss and ds DNA coated tissue culture plates. For each experiment, three normal human sera incubated with biotinylated goat anti-human Ig (H&L) served as negative control. A known positive serum was included in each test. Seventeen sera from SLE patients and five normal sera were tested for reactivity against peptides NG, G, GC, C and H 1. No reactivity was seen when these sera were tested on plates coated with ovalbumin. Fifteen sera from SLE patients and five normal human sera were tested for reactivity to peptides C, GC and Hl(204-218) alone, these same peptides combined with (AT),, and (AT)6 alone. No reactivity was seen in the latter case. Percent inhibition of antibody binding caused by the presence of (AT), was calculated as follows:

Table 1 Peptides from Calf Thymus H I .

Peptide

Sequence

N NG G GC C HI ( 1-1 6) H l(204-2 IS)

1-33" 1-120 34-120 34-2 10 121-210 1-16b 204-2 1x'

"No1used in this study. 'Peptide comprises pan of ihe N-domain. 'Peptide comprises part of the C-domain.

inhibition = 100 -

O.D. H1 fragment+(AT)6 XlOO O.D. H1 fragment

Calculations were made after subtraction of background values for normal sera (O.D. 0.030-0.090).

Absorption and competition studies For absorption studies, 24-well polystyrene tissue cul-


HISTONE HI EPlTOPE EXPRESSION IN CHROMATIN

I69

ture plates (Falcon 3047) were coated with the RESULTS respective antigens using coated buffer as described above. After washing with PBS-Tween, patient sera CD spectra of histone HI peptides and (AT)6 (dilution 1/10) or MRA-12 (15 pg/ml) were incubated To test the effect of (AT), on the helix formation of overnight in the wells at 4°C. Prior to use, NaN3 0.1 % H1 peptides, increasing amounts of (AT), were added was added to all sera. After absorption the sera were to a constant amount of H1 peptide in low salt buffer stored at 4°C until testing. Control sera were treated in and CD measurements were recorded. The CD specthe same manner using uncoated wells. trum of (AT)6 alone did not change much in terms of For competition studies, *log dilutions of fluid elipticity (i.e. negativity) and position of optically phase antigens (10 pl/well) were added to wells active bands in the concentration range 10-120 pg/ml. coated with antigen; immediately thereafter MRA- 12 Histone fragments Hl(1-16), Hl(204-218) and the C (7-14pg/ml) or sera from SLE patients (dilution peptide show a CD spectrum characteristic of a ran1/40) were added. The test was performed in PBSdom coil (Figure 1). Tween to which 0.1% bovine serum albumin (BSA) When increasing amounts of (AT), were added to and NaN3 0.01% had been added. Incubation was for H1( 1-16), increased negativity starting at 220 nm one hour at 37°C. The further procedure was perfor- (Cotton effect) indicated helix formation. The red med as described above. Percent inhibition was calcushift of the optically active band indicates a structural lated as follows: change (Figure 1A). When the amount of (AT), was increased beyond 60pg/ml elipticity diminished % inhibition = 100 (results not shown). Results obtained with O.D. with liquid phase peptide Hl(204-218) were identical (Figure 1B). When (AT)6 O.D. without liquid phase peptide was combined with the complete C-domain the Cotton effect was even more dramatic (Figure 1C). Elipticity x 100 only decreased when more than 120 pg/ml (AT)6 were added to the C domain (results not shown). Results from similar experiments with the GC pep,'..,

V

C 260

'

2io

,

1

1

240

1

260

200

220

240

260

wavelength, nm The influence of adding various concentrations of oligonucleotide (AT), to histone HI peptides H l ( l - 1 6 ) (A), Hl(204-218) (B) and the C domain (C) on elipticity. m" [ a . ~ . ]millidegrees, : arbitrary units. Measurements were taken in low salt buffer. The CD spectrum of (AT), alone at 60pg/rnl is shown in each set of spectra for ease of comparison (- - - -). (A) 0 ,Hl(l-16) at 60 pg/ml; A,Hl(l-16) at 60pg/ml combined with IOpg (AT),; W, Hl(l-16) at 60pg/ml combined with 6 0 p g (AT),. (B) 0 ,Hl(204-218) at 60pg/ml; A, Hl(204-216) at 6 0 p g/ml combined with 6 0 p g of (AT),. (C) 0 , C peptide at 300pg/ml; A, C at 300pg/ml combined with 6 0 p g (AT),; W, C at 300pg/mI combined with 120pg/ml (AT),. Note the increase in elipticity and wavelength with increasing amounts of (AT), added. Figure 1

170

P. CREEMERS. M. MONESTIER A N D L. BOHM

equimolar amounts of H1( 1-16) were negative. No reaction with any of these sera was seen in unrelated antigens (AT)(,or ovalbumin (OVA). When the monoclonal antibody MRA- 12 was tested for binding to the C peptide combined with (AT), in a 1:2 molar ratio, binding was reduced by approximately half as compared to binding to the same amount of C peptide alone. No such effect was found when C was combined with equimolar amounts of ovalbumin (OVA). No binding of antibody to OVA (Figure 3A) or to (AT), was observed (Figure 3 C ) . Reactivity of anti-HI antibodies with HI peptides High concentrations of MRA-12 bound weakly to when combined with (AT), Hl(204-218). Again, binding was reduced by When sera from SLE patients were tested for reactiv- approximately half when H l(204-218) was combined ity to equimolar amounts of H1 peptides NG, G, GC with an equimolar amount of (AT), (Figure 3B). Surand C in the ELISA assay, all sera were found to react prisingly, no influence on binding of MRA-12 to the with GC and C, and none with the central globular G GC peptide occurred when this fragment was comfragment. Only 8 out of 18 sera reacted with NG bined with (AT), in a 1:2 molar ratio (Figure 3C). (Figure 2). As expected, reactivity to complete H1 Fifteen sera from SLE patients were likewise tested was most pronounced. These results are in agreement for binding to the C, Hl(204-218) and GC peptides with previous finding? and we conclude that our alone and combined with (AT),. Sera were tested at a dilution of 1/20. No binding to (AT)(, was observed at patient population is representative of SLE patients. Reactivity to the GC peptide was slightly higher this concentration. The results are expressed in perthan reactivity to the C domain; end point titrations cent inhibition of binding caused by the combination with 9 sera revealed that in five sera the positive end of the respective H1 fragments with (AT), (see point dilution was higher for GC than for C ( 3 sera, I Materials and Methods). The results are depicted in 'log dilution; 1 serum, 2 'log dilutions; 1 serum, 3 *log Figure 4. As for MRA-12, no reduction in reactivity dilutions). Patient sera also bound weakly to equi- was seen when GC was combined with (AT), in a 1:2 molar amounts of H l(204-2 18); average O.D. values molar ratio. However, the combination (AT)6 with C (IkSD) for 1/20 dilution, 0.260f0.080; for 1/40 o r Hl(204-218) resulted in a reduction of reactivity of dilution, 0.150+0.059, N=10. However, reactions to approximately 50%. As a control NG was also tested:


tide and with complete H 1 were inconclusive. There is a relatively large amount of globular G in these peptides which could interfere in the measurement^'^. The results with isolated N and C terminal fragments indicate however that (AT), rapidly combines with N and C terminal ends to produce a helical structure. We therefore used histone HI fragments bound to (AT), as a model to investigate the epitopes of H1 in chromatin.

1.8

' 0

1.4

d 0 Figure 2 Binding as measured by ELISA of sera from SLE patients at a 1/20 dilution to equimolar amounts of calf thymus H I fragments NG. G. GC and C, and to the intact HI molecule. Each point represents the mean of three separate measurements from a single serum. Horizontal bars indicate the average binding of all sera to each antigen. Broken line indicates the highesi binding observed in S normal human sera.

0

'

'.O '

0

:

. 0

0.6

+

0

L

-

NG

A

0 0 0

G

GC

0

-

0 L 0

0 0

L H1

HISTONE HI EPITOPE EXPRESSION IN CHROMATIN

Neither NG-reactive sera ( N = 8 ) nor NG-non-reactive sera (N=9) showed any change in reactivity after combination of NG with (AT)6 in a 1 :2 molar ratio. The fact that binding to (AT)6 did not reduce the reactivity to GC suggests that antibodies show greater affinity for a determinant on G C than for the C peptide.


Absorption and competition studies MRA-12 was absorbed with solid-phase GC and C peptide, and was subsequently tested on equimolar amounts of GC and C. The results were then compared with the reactivity to these fragments of unabsorbed MRA-12. Absorption with one of these HI fragments resulted in loss of reactivity to the other HI fragment as well, indicating cross reactivity between components on GC and C. Nine sera from SLE patients were similarly absorbed with C or GC before testing on these fragments, and results (not shown) were identical.

171

We therefore performed a number of competition studies in the ELISA assay as detailed in Materials and Methods. The results are depicted in Figure 5A. They clearly show that, when tested on either solid phase GC or C antigens, liquid phase GC peptide displaces the solid phase antigen at a higher rate than liquid phase C peptide. Also, both liquid phase antigens show a greater displacement of reactivity to solid phase C than to solid phase G C peptide. Ten sera from SLE patients were likewise analyzed (dilution 1/40) (Figure 5B). The results obtained with sera from SLE patients are more variable than those obtained with MRA-12; standard deviations of results were overlapping. For the three highest liquid phase antigen concentrations, the difference in displacement rate by liquid phase GC as opposed to liquid phase C was significant (P=0.002-0.001, Student’s t-test). Therefore, a preference for GC is still apparent, indicating that also in human sera there is higher affinity for an epitope present on G C than for epitopes on the C peptide.

A 1.6

I\

C 2.0

1.2

1.6

a

0.E

0 0.4

Figure 3 Binding as measured by ELISA of mouse monoclonal antibody MRA-12 to HI fragments. (A) Binding to the C peptide (*-*)and Controls: C combined with ovalbumin (-0). and binding to ovalbumin alone to C when combined with (AT), in a 1:2 molar ratio (- - -0). (-(.)). (B) Binding to Hl(204-218) (*-*)and to HI(204-218) when combined with equimolar amounts of (AT), Control: - -0). Binding to (AT), alone (-0). (C) Binding to the GC peptide (0-) and to GC when combined with (AT),, in a 1.2 molar ratio. (0Control: Binding to (AT), (V-V)alone.

(o---o).

I72

P. CREEMERS, M. MONESTIER AND L. BOHM

8

-&

QAutoimmunity Downloaded from informahealthcare.com by Chulalongkorn University on 01/08/15 For personal use only.

0

0

0 0

-%-

Q

Figure 4 Percent inhibition (see text) of the antibody reaction in the ELISA of sera from SLE patients when the H I fragment is combined with ( A T h Measurements were taken at a serum dilution of 1/20; each point represents the average of 2-4 measurements from a single serum. Horizontal bars indicate the average percent inhibition for each H I -peptide/oligonucleotide combination.

DISCUSSION CD-spectroscopy is an empirical method which can provide information about the secondary structure of macro molecules. The method has been proven useful especially for following conformational changes in proteins24. Our results confirm that the structure of free histone-HI fragments of the N- and C-domain in a low salt solution is that of a random coil, and that the free oligonucleotide (AT), shows spectral features indicative of a single stranded helical structure stabilized by base stacking2s*26. The spectra obtained for the histone fragments HI(I-I6), Hl(204-218) and the C peptide combined with oligonucleotide (AT), are characterized by an increase of negative elipticity with increasing amounts of (AT), added. When a higher than optimum concentration of (AT), was added, elipticity decreased. Results from similar experiments using the GC peptide and complete H I molecule were inconclusive, which we attribute to the presence of a large relative amount of globular G in these peptides, which causes strong innate elipticity. Increased elipticity at 220 nm wavelengths is an indication that an a-helical structure is induced. Our findings with HI(I-I6), Hl(204-218) and the C peptide are in agreement with previous reports: Similar spectral changes are observed when free random coil

HI in solution is subjected to helix promoting a g e n t ~ ’ C-domains ~. of H I respond to charge neutralization by displaying regular helical segments when analyzed by predictive algorithms”. The combination of H 1 fragments with (AT), was therefore used as a model to study antibody recognition of epitopes on histone HI in chromatin. Using the ELISA assay we found that all human sera which were reactive to HI also bound to fragments GC and C, but not to the G peptide. Approximately one third of human sera reacted also with NG. Reactivity to the complete HI molecule was only slightly higher than reactivity to GC or C. These results are in agreement with previously findings4.’.’’. We found that reactivity to the fragment GC was higher than reactivity to C, and positive endpoint dilutions were also higher for the GC peptide. For the elucidation of antibody responses to histones the development of synthetic polypeptides has been proven helpful?’-”. We did not find any reactivity of human sera to H I ( 1-16), whereas weak reactivity was observed to H l(204-2 18). The monoclonal antibody MR- 12 served as a model for comparison with SLE-sera. This antibody reacts in the ELISA with GC and to a lesser extent to C, but in the Western blot only with GC’. An important and crucial finding in this study is that combination of C

HISTONE H 1 EPITOPE EXPRESSION IN CHROMATIN

80

a

0,

'0, 6C

z

0 -

-


Irn I

40

-z

20 I-

z W

U

a W

n

b 40

20

0.16

pM

LIQUID

0.04

0.01

PHASE A N T I G E N

Figure 5 Inhibition of binding of antibodies to solid phase GC and C peptide by 'log dilutions of liquid phase GC and C peptide. Closed symbols, solid phase GC; open symbols, solid phase C. Diamonds, liquid phase GC; squares, liquid phase C. (A) Results obtained with mouse monoclonal antibody MRA-12. Results are the average of two experiments. (B) Results obtained with sera from patients with SLE. Each point represents the average of 3 measurements from 10 different sera.

or its subunit H l(204-2 18) with the oligonucleotide reduces reactivity with the antibody by approximately 50%, whereas combination of GC with oligonucleotide does not result in a reduction in reactivity. Results with fifteen sera from SLE patients were identical. Since (AT)6 is likely to bind equally well to the C peptide and to the C domain in the GC peptide and since anti-H I autoantibodies compete more successfully against (AT), for binding to GC than for binding to C, these findings suggest that these antibodies have

173

a higher affinity for GC than for C. It is therefore likely that the complete epitope is present on GC, whereas only part of this determinant is found on C. Competitive inhibition experiments showing that GC in liquid phase is a more potent inhibitor than C, and that binding to solid phase C is more readily inhibitable than binding to solid phase GC also support the concept of a higher affinity binding to GC. The immunogenic site is probably located at the junction of the G- and C-domain and is not bound to DNA in chromatin. H1 epitopes resembling the immunogenic site which are attached to DNA in chromatin could explain the immense cross-reactivity seen in the free C fragment. This view is supported by our recent sequencing of the variable regions of the MR- 12 monoclonal antibody'". The second hypervariable region of the MRA-12 heavy chain is characterized by the presence of an unusual stretch of acidic amino acids that could interact with the positively charged, DNA-binding residues of the C-domain of histone H 1 ' O . Our study does not allow us to draw definitive conclusions as to whether the production of anti-HI autoantibodies is triggered by free HI or by H1 bound to chromatin. There is almost no published information on the availability of immunogenic nuclear components during autoimmune syndromes. Nevertheless, our data, indicating that antibody binding to the GC epitope is not inhibited by the (AT), oligonucleotide, are compatible with the view that this epitope is accessible in the nucleus and that chromatin could represent an autoimmunogen during lupus syndromes. Acknowledgements We thank Jean Paul Briand of IBMC, Strassbourg for providing two synthetic human HI peptides and Jens Volker and Patrick Bouic for valuable technical assistance. LB was supported by the Medical Research Council of South Africa and the Foundation for Research and Development (CSIR). MM was supported by grant A126665 from the National Institute of Health.

References I . Hardin JA, Thomas JO. Antibodies to histones in systemic lupus erythematosus: Localization of prominent autoantigens on histones HI and H2B. Proc Nut1 Acad Sci USA 1983; 80: 7 4 1 6 7 4 14 2. Costa 0, Monier J-C. Antihistone antibodies detected by ELISA and Immunoblotting in Systemic Lupus Erythematosus and Rheumatoid Arthritis. J. Rheumatol 1986; 13: 722-725 3. Bustin M, Stollar BD. Immunological relatedness of thymus and liver FI histone subfractions. J Bin/ Chem 1973; 248: 3506-35 I0 4. Costa 0, Tchouatcha-Tchouassorn C, Roux B, Monier J-C. Anti-HI histone antibodies in systemic lupus erythematosus: epitope localization after immunoblotting of chymotrypsindigested H 1. Clin Exp tmmunol 1985; 63: 608-6 I3 5. Gohill J, Frilzler MJ. Antibodies in procainamide-induced and


I74

P. CREEMERS, M. MONESTIER AND L. BOHM

systemic lupus erythematosus bind the C-terminus of histone HI. Mol lniniutinl 1987; 24: 275-285 6. Rubin LR, Waga S. Antihistone antibodies in Systemic Lupus Erythematosus. J Rheumafol 1987; 14: 118-126 7. Monestier M, Fasy TM, Bohm L. Monoclonal anti-histone HI autoantibodies from MRL Ipr/lpr mice. Mol Immunol 1989; 26: 749-758 8. Waga S. Tan EM, Rubin RL. Histones in biological fluids: effect on anti-histone antibody specificities. Arthr Rheumatol 1986; 29: S72 9. Rekvig 0, Hannestad K . Human autoantibodies that react with both cell nuclei and plasma membranes display specificity for the octamer of histones H2A. H2B, H3 and H4 in high salt. J Exp Med 1980; 152: 1720-1733 10. Bradbury EM, Cary PD, Chapman GE. Crane-Robinson C, Danby SE, Boublik M. Conformations and interactions of histone H2A (F2A2, ALK). Eiochem 1975; 14: 1876-1885 1 I . Fasman GD, Chou PY, Adler AJ. Prediction of the conformation of the histones. EiophysJ 1977; 16: 1201-1238 12. Van Helden PD, Strickland WN, van Holt C. The complete amino-acid sequence of Histone H2B from erythrocytes of the adult domestic fowl Gallus domesticus. Eiochim Eiophys A c f a 1982; 703: 17-20 13. Bohm L, Sautiere P, Cary PD, Meader DL. Histone HI structure probed by Staphylococcus aureus V8-proteinase. Eiorhim Eiophys Acra 1988; 956: 224-231 14. Clark DJ, Hill CS, Martin SR, Thomas JO. Alpha-helix in the carboxyl-terminal domains of histones HI and H5. EMEO J. 1988; 7: 69-75 15. Maeder DL, Bohm L. The C-domain in the H I histone is structurally conserved. Eiochim Biophys A c f a 199 I ; 276: 233-238 16. Tan EM ef a / . The 1982 revised criteria for the classification of Systemic Lupus Erythematosus. Arth Rheum 1982; 25: I27 1-1 277 17. Bohm L, Strickland WN, Strickland M, Thwaits BH, van der Westhuizen DR, van Holt C. Purification of the 5 main calf thymus histone fraction by gel exclusion chromatography. FEES-left 1973; 34: 2 17-22 I 18. Hartman PG, Chapman GE, Moss F, Bradbury EM. Studies on the role and mode of operation of the very-lysine-rich histone HI in eucaryote chromatin. The 3 structural regions of histone H 1. Eur J Eiochem 1977: 77: 45-5 1

19. Chapman GE, Hartman PG, Bradbury EM. Studies on the role

20. 21.

22. 23. 24. 25. 26. 27.

28.

29.

30.

and mode of operation of the very-lysine-rich histone HI in eucaryote chromatin. The isolation of the globular and nonglobular regions of thc Histone HI molecule. Eur J Eiorhem 1976; 61: 69-75 Bohm L, Sautiere P, Cary PD. Crane-Robinson C. Precise elimination of the N-terminal domain of histone H I . Eiorhem J 1982; 203: 577-582 Muller S, Coupez M, Briand JP. Antigenic structure of histone H2B. Eiochim Eiophys Ar,to 1985; 827: 235-246 Ohe Y, Hayashi H, lwai K. Human spleen histone HI. Isolation and amino acid sequence of the main variant HI b. J Eiorhem 1986: 100: 359-368 Voller A, Bidwell D, Bartlett A. Enzyme-linked immunosorbent assay. In: Rose NR and Friedman H (eds.) Manual ofClinical Immuni)logy Washington D.C. 1980 359-37 1 Mathews K, van Holde KE. Tools of biochemistry, In: Eiochemistry Redwood City, Ca 94065, Cummings Publishing Cy. Inc. 1989; 205-212 Freifelder D. Physical Biochemismy San Francisco U.S.A., W.H. Frieman & Co. San Francisco, USA 1976; 445-467 Tinoco I, Bustamante C, Maestre MF. The optical activity of nucleic acids and their aggregates. Ann Rev Eiophys Bioeng 1980; 9: 107-141 Gohill J, Cary PD, Couppez M, Fritzler MJ. Antibodies from patients with drug-induced and idiopathic lupus erythematosus react with epitopes restricted to the amino and carboxyl termini of histone. J lmniunol 1985; 135: 3 1 16-3 I2 1 Tuaillon N, Muller S, Pasquali J-L, Bordigoni P, Youinou P, Van Regenmonel MHV. Antibodies from patients with rheumatoid arthritis and juvenile chronic arthritis analyzed with core histone synthetic peptides. Inf Arch Allergv Appl Immunol 1990; 91: 297-304 Muller S, Bonnier D, Thiry M. Van Regenmortel MHV. Reactivity of Autoantibodies in Systehic Lupus Erythematosus with synthetic core histone peptides. Inf Arch Allergy Appl Immunol 1989; 89: 288-296 Monestier M. Variable region genes of anti-histone autoantibodies from a MRL/Mp-lpr/lpr mouse. Eur J Imniunol 1991; 21: 1725-1731

Presence of histone H1 on an active Balbiani ring gene.

Histone H1--DNA interaction. On the mechanism of DNA strands crosslinking by histone H1.

Specificities of IgM and IgG anti-histone H1 autoantibodies in autoimmune mice.

Autoantibodies against Modified Histone Peptides in SLE Patients Are Associated with Disease Activity and Lupus Nephritis.

Autoantibodies to nucleolin cross-react with histone H1 in systemic lupus erythematosus.

Monoclonal antibody with specificity to a conserved epitope in the C-terminal domain of histone H1 variants.

Studies on the interaction of H1 histone with superhelical DNA: characterization of the recognition and binding regions of H1 histones.

A DNA-binding homeodomain in histone H1.

Presence of a highly specific histone H1-like protein in the chromatin of the sperm of the bivalve mollusks.

Glycated-H2A histone is better bound by serum anti-DNA autoantibodies in SLE patients: glycated-histones as likely trigger for SLE?

Distribution of H1 histone in chromatin digested by micrococcal nuclease.

Epitope Fingerprinting for Recognition of the Polyclonal Serum Autoantibodies of Alzheimer's Disease.

Histone H1 and the regulation of transcription by nuclear receptors.

Studies on the role and mode of operation of the very-lysine-rich histone H1 (F1) in eukaryote chromatin. Histone H1 in chromatin and in H1 - DNA complexes.

IRF5 is elevated in childhood-onset SLE and regulated by histone acetyltransferase and histone deacetylase inhibitors.

Histone H1 phosphorylation in breast cancer.

A quantitative analysis of histone H1 in rabbit thymus nuclei.

Structure of histone H1-DNA complex: effect of histone H1 on DNA condensation.

DNA binding of histone H1 is modulated by nucleotides.

dBigH1, a second histone H1 in Drosophila, and the consequences for histone fold nomenclature.

The distribution of histone H1 subfractions in chromatin subunits.

Stabilization of sea urchin flagellar microtubules by histone H1.

The C-domain in the H1 histone is structurally conserved.

Allelic isoforms of the chicken and duck histone H1.a.