Immunological Identity of the Small Subunit of HL-A Antigens and β2-microglobulin and Its Turnover on the Cell Membrane

  • Published on

  • View

  • Download

Embed Size (px)


  • Immunological Identity of the Small Subunit of HL-A Antigens and 2-microglobulin and ItsTurnover on the Cell MembraneAuthor(s): Peter Cresswell, Timothy Springer, Jack L. Strominger, Mervyn J. Turner,Howard M. Grey and Ralph T. KuboSource: Proceedings of the National Academy of Sciences of the United States of America,Vol. 71, No. 5 (May, 1974), pp. 2123-2127Published by: National Academy of SciencesStable URL: .Accessed: 03/05/2014 20:39

    Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .

    .JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact


    National Academy of Sciences is collaborating with JSTOR to digitize, preserve and extend access toProceedings of the National Academy of Sciences of the United States of America.

    This content downloaded from on Sat, 3 May 2014 20:39:02 PMAll use subject to JSTOR Terms and Conditions

  • Proc. Nat. Acad. Sci. USA Vol. 71, No. 5, pp. 2118-2122, May 1974

    Cyclic-AMP-Controlled Oscillations in Suspended Dictyostelium Cells: Their Relation to Morphogenetic Cell Interactions

    (chemotaxis/sline molds/cell aggregation/membrane receptors/dissipative structures)

    GUNTHER GERISCH* AND BENNO HESSt * Friedrich-Miescher-Laboratorium der Max-Planck-Gesellschaft, 74 Tabingen; and tMax-Planck-Institut fAr Erniihrungsphysiologie, 46 Dortmund, Germany

    Communicated by Britton Chance, November 28, 1973

    ABSTRACT Periodic spikes of decreased optical density were recorded in cell suspensions of Dictyostelium discoideum. Spike formation as well as clhanges in the redox state of cytochrome b are facultatively and inde- pendently coupled to an oscillating system which is under developmental control and presumably underlies signal transmission in aggregating cells.

    Cyclic AMP triggers a double response, the slow com- ponent resembling the spikes formed during spontaneous oscillations. The fast component shows characteristics of the chemotactic response to cyclic AMP. The receptor system is suiggested to sense changes of cyclic AMP con- centration in time. Cyclic AMP pulses interact with the oscillating system, resulting in phase shift or suppression of spike formation, and in the induction of oscillations in an early stage of development before the onset of spon- taneous oscillations. Continuous flow application of cyclic AMP does not change frequency up to flow rates which extinguish oscillations.

    After the end of growth, cells of the slime mold Dictyostelium discoideum aggregate in response to chernotactic stimuli. This process is an example of self-organization of spatial patterns by chemical cell communication, starting with a layer of randomly distributed identical cells (1-3). Aggregation terri- tories are controlled by centers which typically release chemo- tactic signals in pulses with a frequency of 0.2-0.3 min-' (4, 5). The cells around a center respond by orientated cell movement, and also by producing a pulse to which the outer neighboring cells respond after a signal input/output delay of ) 15 sec. So waves of chemotactic pulses can be propagated over a distance much larger than the chemotactic action radius of an aggregation center (4, 5).

    cAMP elicits a chemotactic response (6, 7), and when ap- plied in pulses induces propagated waves, thus simulating transmitter action (8). Extracellular cAl\IP is rapidly de- stroyed by extracellular as well as cell-bound phosphodi- esterases (9, 10). Periodic activities and cAMP effects can be recorded optically in stirred cell suspensions. This makes it possible to investigate the molecular basis of morphogenetic cell communication under conditions similar to those used in studying oscillations of the glycolytic pathway in yeast cell suspensions (11, 12). In the present paper we report that the ability to oscillate in suspensions is related to the morpho- genetic capacity of the cells, and describe interactions of cAMP with the oscillating system.

    Abbreviations: cAMP, cyclic adenosine 3',5'-monophosphate; cGMP, cyclic guanosine 3',5'-monophosphate; cTMP, cyclic inosine 3', '-monophosphate; t,, developmental stage of cells timed as n hours after their removal from growth medium.


    Dictyostelium discoideum strain Ax-2 (13) was cultivated at 22-25? axenically on growth medium containing 1.8% mal- tose (13) up to cell densities of 0.3 to 1.4. 107/ml. The cells were washed three times in the cold with 0.017 M Soerensen phosphate buffer, pH 6.0, resuspended in the buffer, adjusted to 1- 107/ml, and shaken. The time of resuspension was taken as the beginning of cell differentiation to aggregation com- petence. After various times, cells were centrifuged and ad- justed in cold buffer to 2- 108/ml. From an ice bath, 2 ml of the suspension were transferred into a cuvette with an optical pathway of 1 cm, and agitated by bubbling oxygen with a constant flow rate of 24 4+ 1 nIl/mii through two syringes. For all measurements taken at 405 and 430 nm, the cuvette was kept at 230. The spectrophotometer and the continuous flow equipment used are described in (14) and (12), respec- tively.

    Recording of cytochrome b was based on oxygen-dithionite difference spectra determined at liquid air temperature in suspensions of 108 cells per ml using a Johnson Foundation split beam spectrophotometer. Peaks were found for cyto- chromes a, a3 at 598, cl at 553, c at 548, and b at 562 and 560, as well as a Soret region with a peak at 425 and shoulders at 430 and 445 nm. The redox state of cytochrome b in living cell suspensions was recorded at 430 nm using 405 nm as the ref- erence wavelength. The latter was simultaneously used for recording optical density.


    Spikes and Sinusoidal Oscillations. After separation from the growth medium, suspended cells of D. discoideum Ax-2 pass through a pre-aggregation phase of about 9 hr before they acquire full aggregation competence (9, 15). Within the first 5 hr of this phase, no spontaneous oscillations were ob- served. When, however, cells were harvested 6-14 hr after separation from the growth medium, oscillations began im- mediately after transfer from an ice bath into the optical cuvette. Regularly, an initial series of spikes was recorded, followed by sinusoidal oscillations (Fig. 1). Sometimes several cycles of only sinusoidal oscillations were intercalated be tween spike-generating cveles. The mean spike frequency was 0.14 min-', and the frequency increased upon cessation of spike forrimation, with a mean acceleration factor of 1.20. These results indicate that the cellular activity underlying spike formation, although being coupled to an oscillating system, is not an indispensable part of it.


    This content downloaded from on Sat, 3 May 2014 20:39:02 PMAll use subject to JSTOR Terms and Conditions

  • 2124 Immunology: Cresswell et al. Proc. Nat. Acad. Sci. USA 71 (1974)


    80 - A-Microglobulin

    ~60- S

    / L-~~~~~~~~A7,12 40-


    I I 1 1 Is { ,,1 0.001 0.01 0.1

    pg Protein

    FIG. 1. Inhibition of precipitation of ['2511f.32-microglobulin by rabbit antiserum to #g-microglobulin with ,B2-microglobulin purified from human urine or with HLA7,12 antigen prepared after treatment of membranes of cultured human lymphoblastoid cells (RPM1 4265) with papain.

    the HL-A antigen preparations by studying the capacity of these antigens to inhibit the cytotoxicity reaction (Fig. 2B). By this method it was estimated that for the papain-solubi- lized produicts 30% of the HL-A2 antigen preparation or 39% of the HL-A7,12 antigen preparation was composed of mate- rial crossreactive with fl2-microglobulin (Table 1). In a similar experiment with detergent-solubilized HL-A antigen prepara- tions which were about 50% pure (3, 4) 12% of the inhibitorv capacity of purified ,32-microglobulin was obtained, which, taking into account the larger molecular weight of the deter- gent-solubilized material and its degree of purity, closely correlates with the inhibition figures obtained with the papain- solubilized material. The data obtained with these two sepa-

    rate inhibition studies, although giving somewhat different values for the amount of 32-mnicroglobulin present in HL-A antigen preparations, clearly indicate that a large amount of 32-microglobulin was present in these preparations (23- 39%), considerably more than could be accounted for by contamination with an unrelated protein. This conclusion is strengthened by the fact that the preparative procedures for the two HL-A antigens isolated after papain digestion in- volved their elution at different salt concentrations in ion exchange chromatography (1) and that one of the steps in the purification of the detergent-solubilized material (4) was puri- fication of glycoproteins by passage over a lectin column which would not bind a carbohydrate-free protein such as A32-micro- globulin unless it was bound to carbohydrate-containing molecules. The values obtained are in good agreement with the 35% by weight of the 12,000-MW polypeptide in papain- solubilized HL-A, antigen calculated on the ratio of [5H]- aminoacids in the two polypeptides (1). The cytotoxicity assay measures only the subset of #32-microglobulin antibodies directed at determinants exposed in the cell-bound form of 02-microglobulin, yet approximately the same proportion of 32-Microglobulin was found in HL-A antigen by this assay as

    by the assay employing inhibition of [1251 ]32-microglobulin precipitation assay. Therefore, most of the antigenic determi- nants of soluble 02-microglobulin are also found in its cell- bound form.

    Thus, quantitatively by two different methods, the small peptide of HL-A antigens was equally as effective as 32- microglobulin in combining with 02-microglobulin antisera. These data indicating virtually total cross reactivity appear to exclude the possibility that the 12,000-molecular-weight

    70 B

    A BSA - a 2 Microglobulin

    x Papoih soluble HL-A2 * Papain soluble HL-A7+12

    dJ) D0 Detergent soluble cu 0HL-A2,7,12


    ~~~~~30 ~ ~ ~ ~ ~ 0

    Antiserum dilution-1

    10C, I , I, .,,, I I -I I,,,, I

    0 ) 5 10 50 100

    Antigen Dilution' FIG. 2. Lysis of human peripheral lymphocytes by rabbit antiserum to 32-microglobulin and inhibition of the lysis by HL-A and by

    32-microglobulin. (A) Lysis. The rabbit 32-microglobulin antiserum was diluted serially 2-fold in 10 IAd of dextrose-gelatin-Veronal-saline (DGV) plus 1:40 normal rabbit serum. Peripheral human lymphocytes (TS, HL-A ..... ) (5 JAl containing 3 X 106 cells per ml, labeled with 51Cr) were added and incubated 1/2 hr at 37?. Rabbit complement, diluted 1:4 with 0.14 M NaCl, 0.01 M Tris-glycine pH 8.3, 1 mM MgCl2, 0.3 mM CaC12, 0.01 mM ethylenediaminetetraacetate (EDTA), 100 Il, was added and the mixture incubated 45 min at 37?. Lysis was stopped by addition of 50 ,ul of 20 mM EDTA, 0.14 M NaCl, 0.01 M sodium phosphate, pH 6.8. After centrifugation at 1000 X g for 5 min, 50 .dA of the supernatant solution was removed for 5'Cr counting. (B) Inhibition of lysis. Antigens were diluted in DGV plus 1:40 normal rabbit serum before assay. The amounts used are shown in Table 1. Detergent-soluble HL-A (4) which contained 0.34% Brij 99 (5 ,l) was diluted with an equal volume of 30% bovine-serum albumin (BSA) before use. A sample containing BSA alone was also titrated and found to be completely noninhibitory. Antigens were diluted 2-fold serially through 1:40 32-microglobulin antiserum in D)GV. After incubation for 1 hr at 370, the assay was continued as described above.

    This content downloaded from on Sat, 3 May 2014 20:39:02 PMAll use subject to JSTOR Terms and Conditions

  • Proc. Nat. Acad. Sci. USA 71 (1974) #2-Microglobulin and HL-A Antigens 2125

    4 6 43j

    FIG. 3. Double diffusion of different HLA preparations with antisera to 8j-microglobulin. (A) Center well, 12 ul of rabbit antiserum to 62rmicroglobulin. Wells 1 and 4, 0.5 ,ug of 62-micro- globulin; well 2, 3 MAg of papain-solubilized HIA7; well 3, 1.5 Mg of same; well 5, 1.5 MAg of papain-solubilized HLA2; well 6, 3 Mug of same. (B) Center well, 12,ul of turkey antiserum to #2-micro- globulin. Wells 1-6 as in A, above. (C) Center well, 12 ul of turkey antiserum to r-microglobulin. Wells 1 and 5, 0.5 Mg of l2- microglobulin; well 2, 1.5 MAg of papain-solubilized HL-A7; well 3, empty; well 4, 4.5 Mg of papain-treated detergent-solubilized HL-A2,7,12; well 6, 6.5 MAg of detergent-solubilized HLA2,7,12; well 7, 0.5 Mg of papain-treated ,2-microglobulin; well 8, 1.5 ug of papain-solubilized HI-A2. (D) Center well, 12 Ml of rabbit anti- serum to ,B2-microglobulin; wells 1-8 as in C above. (The turkey antiserum was the generous gift of Mr. Harvey Faber, University of Wisconsin.)

    subunit of HL-A antigen is a closely related polypeptide which cross reacts with #2-microglobulin. It is noteworthy that, despite extensive homology, antisera to #2rmicroglobulin do not cross react with immunoglobulins or vice versa.

    Immunological Identity of the Reaction of Antisera to 02- Microglobulin with HL-A Antigens and #2-MIicroglobulin. The reactions of both rabbit and turkey antisera to #2-microglob- ulin with HL-A antigens prepared after papain treatment or after detergent solubilization, and with j32-microglobulin were examined by Ouchterlony double diffusion in agar. Lines of complete identity were obtained in all cases with no evidence of spurring (Fig. 3).

    Identity of the Small Subunit of HL-A and 132-Microglobulin by SDS and SDS-Urea Gel Electrophoresis and by Isoelectric Focusing. The various preparations of detergent- and papain- solubilized HL-A antigens and #2-microglobulin were sub- jected to gel electrophoresis alone or together (Fig. 4A and B). In all cases the 12,000-molecular-weight peptides were found to be identical. On isoelectric focusing in 7.5% polyacrylamide gels containing 1% Ampholine, pH 4-6, 2-microglobulin showed a single band having pl 5.2. Each of the HL-A prepa- rations yielded a band in an identical position which on elu- tion and SDS-gel electrophoresis had a molecular weight of

    A. 1 2 3 B. 1 2 3 4 5 6


    12,000- FIG. 4. Identity of 6r-microglobulin and the small peptide in

    various HI-A preparations. (A) SDS-gel electrophoresis (Laemmli SDS gels with 12% acrylamide)....