Comparison of MHC antigen expression on PHA- and MLC-induced T cell lines with that on T and B lymphoblastoid cell lines by cell cycle depedency

Comparison of MHC Antigen Expression on PHA- and MLC-Induced T Cell Lines with That on T and B Lymphoblastoid Cell Lines by Cell Cycle Dependency* Yoshiki Matsui, Donald E. Staunton, Howard M. Shapiro, and Edmond J. Yunis

ABSTRACT: Although it is well known that the expression of major histocompatibility complex (MHC) antigens on the surface of lymphoblastoid cell lines are cell cycle dependent, the way in which the M H C antigen expression on activated T cells varies with cell cyclephase has not previously been described. Using 11 lymphoblastoid cell lines from malignant and nonmalignant tissues (B cells, T cells, and myeloid cells) andfive activated T cell lines (two cell lines activated by phytohemagglutinin and three alloreactive T cell clones), MHC antigen expression was quantitatively studied by dual-beam flow cytometry. Correlated measurements of surface antigen quantity (immunofluorescence), D N A content (Hoechst 33342), and cell size (light scatter), uninfluenced by induction synchrony and cellfixation, were performed. The data indicate that cell surface antigen quantity and cell surface area demonstrate specific values at each phase of the cell cycle when the cells are in logarithmic growth. Examining cells in logarithmic growth, it was confirmed, for all lymphoblastoid cell lines, that the quantity of MHC antigens on G2 (S + G~ + M) cells was greater than that on G1 cells. In addition, it was found, by analyzing antigen quantity and surface area, that class I antigen density in the G2 phase is 17% less than that in the G~ phase in leukemic T cell lines, and that both class I and class II antigen densities in the G2 phase were 21% less than that in the G1 phase in lymphoblastoid B cell lines. In activated T cells, class ! antigen density in the G2 phase was 11% less than that in the G: phase, while class II antigen density in the G2 phase was 12% greater than that in the G1 phase. We describefour important observations in this report. In both G1 and G2 phases, activated T cells express: (1) quantitatively fewer class I antigens than lymphoblastoid B cell lines; (2) similar quantity of class I antigens as that of leukemic T cell lines; and (3) similar quantity of class II antigens as that of lymphoblastoid B cell lines. Also, (4) class II antigens are expressed in greater density in the G2 phase than in the G~ phase in activated T cells. In contrast, lymphoblastoid B cell lines express greater density of class II antigens in the G1 phase than in the G2 phase of the cell cycle. These findings differ from previous reports, suggesting that G1 phase cells may have a more significant role than G2 phase cells as target cells for MHC restricted cytotoxic cells.

From the Division of lmmunogenetics, Dana-Farber Cancer Institute; the American Red Cross Blood Services-NortheastRegion; the Centerfor BloodResearch;and the Departmentof Pathology, Harvard Medical School, Boston, Massachusetts. *This work was supportedin part by grantsfrom the National Institutes of Health CA06516, CA20531 and All 7120, and a grant from the American National Red Cross Matching Funds. Address reprint requests to Yoshiki Matsui, M.D., Division of lmmunogenetics, Dana-FarberCancer Institute, 44 Binney Street, Boston, MA 02115. ReceivedMay 13, 1985; acceptedAugust 2, 1985. Human Immunology 15,285-301 (1986) © Elsevier Science Publishing Co., Inc., 1986 52 Vanderbilt Ave., New York, NY 10017

285 0198-8859/86/$3.50

286

Y. Matsui et al.

ABBREVIATIONS

CTL EBV FCS HBSS H0342

cytotoxic T lymphocytes Epstein-Barr virus fetal calf serum Hanks' balanced salt solution the DNA dye Hoechst 33342

MHC MoAb PHA PHA-CM

major histocompatibility complex monoclonal antibody phytohemagglutinin PHA-induced conditioned medium

INTRODUCTION The major histocompatibility complex (MHC) has been shown to play an important role in the regulation of immune responses. Recently, it has been reported that quantitative variations in the cell surface expression of MHC gene products influence cell-cell interactions [1-6]. The recognition of virus-induced tumor antigen with MHC class I antigens by tumor-specific cytolytic T lymphocytes (CTL) is correlated with the quantitative expression of class I and virus-induced antigens [1]. MHC class II (Ia) antigen density on B cells [2,3] and B lymphoma cells [4] is also critical in T-B cell interactions. The magnitude of T cell proliferative responses is a function of the amount of class II antigen expressed on antigen-presenting cells [5,6]. If the density of MHC antigens on the cell surface differs with position in the cell cycle, then the cell cycle distribution of a population of cells may influence the result of cell-cell interaction. It would thus be important to know whether the expression of class II antigens on activated T cells fluctuates during the cell cycle, because activated T cells, which express class II antigens, can function as antigen-presenting cells and also as allogeneic stimulator cells [7]. In previous studies using very few cell lines, the expression of class I and class II antigens on the surface of T and B lymphoblastoid cell lines has been determined using synchrony induction methods or fixation of cells [8-12]. It was reported that synchrony induction methods do not reflect the true cell cycle distribution, arresting DNA synthesis but not enzyme activity, and that both synchrony induction methods and cell fixation reduce cell size [ 11,13-15]. Despite such possible limitations of these studies, it has been shown that there is cell cycle dependency in expression of class I and class II MHC antigens, and that the density of MHC antigens in the G2 phase is either the same as or greater than that in the G1 phase [8-12]. Although it is known that class I and class II antigens are expressed on activated T cells [16,17], it is not known whether the expression of MHC antigens on these cells is cell cycle dependent, and, if so, whether the dependency is similar to that of T or B lymphoid cell lines. In order to investigate these points, we have examined the cell cycle dependency of MHC antigen expression in 16 cell lines, including phytohemagglutinin (PHA)-stimulated T cells, T cell clones with cytotoxic and/or natural killer-like functions, leukemic T cells lines, Epstein-Barr virus (EBV) transformed lymphoblastoid B cell lines, and Burkitt's lymphoma cell lines. Dual-beam flow cytometry, which offers a precise and direct method for relative quantitation of cell size, surface immunofluorescence, and DNA content, was used to make correlated measurements of these values. The determination of cell cycle dependency on antigen density made by this method is not influenced by induction synchrony and/or cell fixation. The main finding of the study is that the G2/G1 ratio for both quantity and density of class I antigens on activated T cells was significantly greater than that on either leukemic T cell lines or lymphoblastoid B cell lines, and the G2/GI ratio for both quantity and density of class II antigens on activated T cells was significantly greater than that on B lymphoblastoid cell lines. In activated T cells, the antigen density of class

Cell Cycle Dependency of MHC Antigen Expression

287

I antigens was greater in the G, phase, while that of class II was greater in the G2 phase of the cell cycle. MATERIALS AND METHODS

Cell lines. Sixteen human cell lines were used in the present study: CEM [18], H-SB2 [19], MOLT-4 [20], HPB-ALL [21], and Jurkat [22] are human leukemic T cell lines; Daudi [23] is a Burkitt's lymphoma cell line with B cell characteristics; SB is a lymphoblastoid B cell line established from the same acute lymphocytic leukemia patient from whom H-SB2 was established [19]; Laz007 is a lymphoblastoid B cell line established from lymphocytes of a normal individual by transformation with EBV [24]; HL-60 is a myeloid leukemic cell line [25]. These cell lines were kindly provided by Dr. J. Minowada, Roswell Park Memorial Institute, Buffalo, NY and by Dr. H. Lazarus, Dana-Farber Cancer Institute, Boston, MA. PHA-T1 and -T2 are T cells activated by PHA and maintained with conditioned medium derived from PHA-stimulated human lymphocytes. Clones 9, 18, and 23, derived from a single normal donor, are alloreactive human T cell clones with cytotoxic, natural killer-like and noncytotoxic functions, respectively [26]. EBV-B1 and -B2 are lymphoblastoid B cell lines transformed by EBV. These cell lines were established in our laboratory from normal healthy donors. Cells were maintained in suspension culture in RPMI 1640 medium containing 10% heat-inactivated fetal calf serum or pooled human serum with 2 mM supplemental L-glutamine, 100 IU/ml penicillin, and 100 ~g/ml streptomycin. Cultures were kept in 75 cm ~ tissue culture flasks at 37°C in a humidified atmosphere containing 5% CO2 with twice weekly passage, splitting and feeding. Culture media were changed and cell concentrations were adjusted to 2 x 105 cells/ml on the day before assays.

Monoclonalantibodies.

A murine monoclonal antibody (MoAb) specific for human MHC class I (HLA-A,B,C) heavy chain, W6/32, was purchased from Pel-Freez Biologicals, Rogers, AR [27], and a murine MoAb recognizing human MHC class II (HLA-DR) antigens, I-2, was kindly provided by Dr. L. Nadler, DanaFarber Cancer Institute, Boston, MA [28].

Staining procedures. Cultured cells were washed once in warm medium with special attention given to dispersing clumped cells after both centrifugations. The DNA dye Hoechst 33342 (H0342) (Hoechst AG, Frankfurt, Federal Republic of Germany) was added from 100 ~M solution to 106 cells/ml in RPMI 1640 to a final concentration of 51zM. At 5 /zM, H0342 has previously been shown to be a stoichiometric stain for DNA [29-31]. Cells were incubated at 37°C for 1 hr, and then pelleted, washed twice in ice cold medium, and resuspended in a saturating concentration of MoAb in Hanks' balanced salt solution containing 5% fetal calf serum (HBSS/5% FCS). After 30 min on ice, they were washed three times with cold HBSS/5% FCS and incubated for an additional 30 min with 30 ~l of a 1:40 dilution of the fluorescein-conjugated F(ab')2 fragment of rabbit anti-mouse IgG (Cappel Laboratories, Inc., Cochranville, PA). The samples were washed three times and maintained at 0-4°C until flow cytometric analysis.

Flow cytometry. The dual illumination beam instrument used in this study has been described [32-34]. Briefly, the blue fluorescence of H0342 was excited by uv light (350-370 nm) from a mercury arc lamp; the emitted fluorescence was detected by a photomultiplier fitted with a 420 nm long-pass filter and a 450 nm broad-band interference filter. The green fluorescence (530-570 nm) of fluo-

288

Y. Matsui et al. rescein was excited by the 488 nm beam o f an argon ion laser. The 488 nm beam was used to make m e a s u r e m e n t s of forward light scatter (cell size). Both histograms and two-parameter distributions were accumulated using hardwired analyzers incorporating storage display oscilloscopes.

Data analysis. T h e mean intensity of immunofluorescence and light scatter per cell were c o m p u t e d by using the formulas (F × N ) / ~ N and ~(S x M)/~ M, respectively, where F and S are fluorescence and light scatter channel numbers, and N and M are numbers of cells in the F and S channel, respectively. W h e n the same suspension o f labeled cells was assessed twice, less than 3 % difference in the c o m p u t e d mean was found. Most of the determinations were performed singly and the data in the tables appear without the standard error of the mean. The intensity o f the green fluorescence signal generated for each cell passing through the flow cytometer is linearly related to the n u m b e r of fluorescein molecules present on the cell surface [35]. Therefore, the G2/G1 ratio o f antigen quantity on the cell surface is equal to the G2/G, ratio of the fluorescence intensity. The G2/G, ratio of the cell surface

Dual-beam flow cytometric analyses of DNA content, antigen expression, and cell size. HPB-ALL (I) and clone 18 (II) were labeled with both DNA dye (H0342) and immunofluorescence. Two-dimensional distributions of DNA content (H0342 fluorescence) and MHC antigen expression (immunofluorescence) (Ia, IIa), and that of DNA content and cell size light scatter (Id,IId) were obtained using the flow cytometer. Histograms of the class I antigen expression on HPB-ALL in the G, phase (Ib) and in the G2 (S + G2 + M) phase (Ic), and that of the class II antigen expression on clone 18 in the G1 phase (IIb) and in the G2 phase (IIc), were obtained using windows on the display defining the cell cycle position. Histograms of cell size in the G~ phase (Ie) and in the G2 phase (If) of HPB-ALL, and that in the G1 phase (lie) and in the G2 phase (IIf) of clone 18 were also observed using the windows. For each cell line, 10,000 cells were analyzed. FIGURE 1

Ia

Ib

Ic

G 1G 2 S÷M

G~

G2 S+M

G1 G2 S+M

G,

G2 S+M

Id

Ie

If

IId

IIe

IIf

G~ G 2 S+M

G,

G2 S+M

G, G2 S'~M

GI

G2 S÷M

HO342 Fluorescence P

lla

Cell Number b

lib llc

HO342 Fluorescence ~

Cell Number i

Cell Cycle Dependency of MHC Antigen Expression

289

area (~" d 2) was calculated: (i) by measuring the light scatter intensity of cells in G1 and G2 phases; (ii) by using calibration plots of light scatter of Microspheres (9.57, 14.62, and 20.13/zm in diameter, Coulter Electronics, Inc., Hialeah, FL); (iii) and by using the formula I = k(d) n [36], where I is light scatter intensity, k and n are constant, and d is diameter. The antigen density on the cell surface was obtained by dividing the surface antigen quantity by the cell surface area. Comparisons were determined by use of the two-tailed Student's t test.

RESULTS Correlated Measurements of D N A Content, Surface Antigen Quantity, and Cell Size Two-dimensional distributions of H0342 fluorescence (DNA content) and immunofluorescence (cell surface MHC antigens), and of H0342 fluorescence and light scatter (cell size), were determined in 16 cell lines; representative results are illustrated in Figure 1. A single dense cluster of cells in the G1 phase and a single coarse cluster of cells of the S + G2 + M phases, were observed in the two-dimensional distributions of DNA content and antigen expression, and of DNA content and cell size (Figure 1). Cells were classified into two groups, Gl and G2 (S + G2 + M), according to the magnitude of H0342 staining, because the S and M phase cells constituted small populations which were identical or close to the G2 phase cells with respect to antigen quantity and cell size. Thus, the fluctuation of antigen density during the cell cycle was determined considering both the quantity of antigen and the cell size in the G1 and G2 (S + G2 + M) phase. For example, the immunofluorescence intensities (the quantities) of MHC class I antigens expressed by HPB-ALL in the G1 and G2 phases were almost identical [Figure l(Ia),(Ib),(Ic)]. However, class II antigen quantity on clone 18 in the G2 phase was much greater than that in the GI phase [(Figure l(IIa),(IIb),(IIc)]. The light scatter (cell size) of both HPB-ALL and clone 18 in the G2 phase was greater than that in the G1 phase [Figure l(Id),(Ie),(If),(IId),(IIe),(IIf)]. Comparison of M H C A n t i g e n Quantity Between G1 and G2 Phases The relative quantities of the surface antigen on cells of each cell line (i.e., the mean immunofluorescence intensity obtained from the respective histograms) are shown in Figures 2 and 3. The relative quantity of both class I and class II antigens in the G2 phase was greater than that in the G1 phase for all the cell lines examined. The mean quantity of class I antigens on the surface of lymphoblastoid B cell lines in the G1 phase was 235 -+ 12 (mean channel number +SEM) (Figure 2). This mean was significantly higher than that of activated T cells (p < 0.001), or of leukemic T cell lines (p < 0.009). The mean quantity of class I antigens on the surface of lymphoblastoid B cell lines in the G2 phase was 304 _+ 14. This value was significantly higher than that for activated T cells (p < 0.05), or that of leukemic T cell lines (p < 0.02). No significant difference in the mean quantity of class I antigens between activated T cells and leukemic T cell lines was observed (Figure 2). No significant difference in the mean quantity of class II antigens between activated T cells and lymphoblastoid B cell lines was observed (Figure 3). The G2/G1 ratios of antigen quantity for both class I and class II antigens can be considered as independent variables because this ratio showed no significant correlation with antigen quantity in either the G1 or G2 phases (p > 0.1). In the present study, as in previous works, class I antigen was not detected on Daudi cells [37], and class II antigen was detected neither on

290

Y. Matsui et al.

CELL

LINE

PHA-T

1 L . . . . . . . . u e . . . . . . . . J. . . . . . . .

PHA-T2

Clone Clone

[. . . . . . . . . . . . . . . . . . . . . .

I ................

9 L 18

I

[~-. . . . . . . . . . . . . .

~

Clone 23 [ M e a n + _ S E M ~.

...... -J. . . . . . . . . . . . --r,...__au__,

C E M ~. . . . . . . . . . . . . . . . . . . . . . . H-SB2

[. . . . . . . . . . . . . . . . . . . .

MOLT4

[. . . . . . . . . ) . . . .

HPB-ALL

L. . . . . . . . . . . . . . .

Jurkat Mean*

EBV-B

/-. . . . . . . . 3. . . . . . .

J.._

L ..................

SEM

_t . . . . . . .

~=m=m~E=dm~

1 L .........................

EBV-B2

P
P < 0.05(G 2)

I

P <0,009(G0 P < 0.02(G2)

.*. . . . . . . . . .

[-- . . . . . . . . . . . . . . . . . . . . . . .

I .......

L a z 0 0 7 [. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SB [

) ..........

....,l.

M e a n :S:8 E M [

§ (G0 §S (G2)

!

HL-60

L................

I

I

J. . . .

I

I

I

0

100 200 300 400 Quantity of Class I Antigen (Channel Number)

FIGURE 2 Comparisons of the relative quantity of MHC class I antigens on the surface of the cell lines. The relative quantity of the surface antigens, or the means of the immunofluorescence intensity, in the G1 phase (open bars) and in the G2 phase (hashed bars) were obtained from the histograms of the class I antigen expression using the flow cytometer and the windows defining the cell cycle position. Representative results from repeat studies using the same cell lines are shown. The mean quantities of class I antigens on the cell surface of lymphoblastoid B cell lines (EBV-B 1 and -B2, Laz 007, and SB) in the G1 (§) and G2 (§§) phases were significantly greater than that of activated T cells (PHA-T1 and -T2, and clones 9, 18, and 23) in the G1 and G2 phases, respectively (p < 0.001 and p < 0.05), and also significantly greater than that of leukemic T cell lines (CEM, H-SB2, MOLT4, HPB-ALL, and Jurkat) in the GI and G2 phases, respectively (p < 0.009 and p < 0.02). Daudi was negative for class I antigens.

any of the leukemic T cell lines [21,38], nor on HL-60 [39] (data not shown). T h e immunofluorescence intensity o f cells in both G1 and G2 phases stained with the fluorescein-conjugated anti-mouse I g G antibody alone, was very low (> 2 channel number), and differences between cells in the G] and G2 phases were not significant (data not shown). C o m p a r i s o n o f Cell S u r f a c e A r e a s B e t w e e n G1 a n d G2 Phases T h e cell surface area in the G2 phase was 51 to 8 6 % greater than that in the G , phase (Table 1). N o significant difference in the means of the G2/G1 ratios between the groups o f cell lines were observed (range o f means 1.60-1.68). N o significant difference was observed in cell size with or without immunofluorescence staining (data not shown).

291 CELL LINE

/ P H A - T 1 ~.,-~,~-~-.,,-,-/,-~-,,~-~,~-,-/,-~-.r~-)////////////A Clone 9 [////HHHHHH,'~/H/HHHH,',W

°,o°e

ZZ;

....

Clone 23 Mean ± SEM L. . . . . . . . . . . . . . . . . . ~

................

:

I

NS(GI)

NS(G2)

EBV-B1 L....................... ; ..... EBV-B2 [

I

I

Lez 0 0 7 [

. . . . . . . . . . . . . ). . . . . . . . . . . . . .

/

SB L~. . . . . . . . . . . . . . . . . . . . . .

I

Mean + 8EM I,

[

0

,). . . . . . . . . .....I~..=....~

,

!

I

o (GO

Oo ( G 2)

,

100 200 300 Quantity of Class II Antigen

(Channel Number) FIGURE 3 Comparisons of the relative quantity of MHC class II antigens on the surface of the cell lines. The relative quantity of the surface antigens, or the means of the immunofluorescence intensity, in the G; phase (open bars) and in the G2 phase (hashed bars) were obtained as described in the legend of Figure 2. Differences in the mean quantities of class II antigens on the cell surface between activated T cells (PHA-T1, and clones 9, 18, and 23) in the G] and G2 phases and lymphoblastoid B cell lines (EBV-B1 and -B2, Laz 007, Daudi, and SB) in the G1 ( 0 ) and G2 ( O O ) phases were not significant, respectively. HPA-T2 was not tested. All leukemic T cell lines and HL-60 were negative for class II antigens.

TABLE 1

Comparison o f cell surface area between GI and G2 phase in cell lines G2/G1 ratio of cell surface area

Activated T cells PHA-T1 1.60 PHA-T2 1.80 Clone 9 1.66 Clone 18 1.60 Clone 23 1.75 Mean ± SEM 1.68 - 0.04 Leukemic T cell lines CEM 1.61 H-SB2 1.51 MOLT 4 1.72 HPB-ALL 1.51 Jurkat 1.66 Mean ± SEM 1.60 ± 0.04

Lymphoblastoid B cell lines EBV-B1 1.55 EBV-B2 1.54 Laz 007 1.85 Daudi 1.67 SB 1,63 Mean ± SEM 1.65 ± 0.06

HL-60

Myeloid leukemic cell lines 1.86

292

Y. Matsui et al.

Although both antigen quantity and surface area of cells are greater in the G2 phase than the G1 phase in all cell lines tested, antigen density (antigen quantity/unit surface area) was not constant during the cell cycle (Table 1, Figures 2 and 3). No significant correlation between the G2/G1 ratio of surface area and the G2/G; ratio of antigen quantity was observed for either class I or class II antigens. Therefore, the G2/G~ ratio of antigen density, obtained by division of the G2/GI ratio of antigen quantity by the G2/G1 ratio of surface area, can be considered as an independent variable. C o m p a r i s o n of M H C A n t i g e n Density B e t w e e n G1 and G2 Phases The mean G2/G; ratio of the relative quantity of class I antigens on the cell surface of activated T cells was 1.50 - 0.06 (mean _ SEM) [Figure 4(a)]. This mean ratio was significantly higher than that for leukemic T cell lines, which was 1.33 - 0.03 (p < 0.04), or that for lymphoblastoid B cell lines, which was 1.29 ± 0.02 (p < 0.03) [Figure 4(a)]. The mean G2/G~ ratio of the class I antigen density on the cell surface of activated T cells was 0.89 ± 0.02 [Figure 4 (b)]. This mean ration was significantly higher than that for leukemic T cell lines, which was 0.83 ± 0.01 (p < 0.02), or that for lymphoblastoid B cell lines, which was

F I G U R E 4 The G2/G1 ratio of MHC class I antigen expression on the surface of the cell lines. The G2/G] ratio of the surface antigen quantity (4a) was obtained by dividing the relative quantity of the surface antigens in the G2 phase by that in the GI phase. The G2/G1 ratio of the surface antigen density (4b) was obtained by dividing the G2/GI ratio of the surface antigen quantity by the G2/G, ratio of the cell surface area shown in Table 1. The means of the G2/G~ ratios of antigen quantity (*) and antigen density (**) in activated T cells were significantly greater than that in leukemic T cell lines (p < 0.04 and p 0.02, respectively) (4a, 4b), and also significantly greater than that in lymphoblastoid B cell lines (p < 0.03 and p < 0.03, respectively) (4a, 4b). CELL

4a

LINE

P H A - T 1 II PHA-T2 I[

I

Clone 9 h Clone 18 II Clone 2,5 It Mean +_8EM I t GEM It H-SB2 h M O L T 4 J| HPB-ALL II Jurket II

4b

I ~L.

I

I I ,

;

:

I I'~

•

ee

J i J J

Mean : i : S E M

J

P
E B V - B 1 )!

i

E B V - B 2 !i Lez 0 0 7 8B

i

II II

j

J

] == :

:P
*-.°

1.2

]

II II

P
P
I

)t

!

Mean * 8 E M

~=nmy 0

I

I

|

1.4

1.6

G21G I R a t i o

Ir----~ II i

0

0.6

of Class

i 0.8 I Antigen

1.0

Density

Cell Cycle Dependency of MHC Antigen Expression

293

0.79 ± 0.04 (p < 0.03) [Figure 4 (b)]. These results indicate that, in all cell lines, the density of M H C class I antigens on the cell surface in the G1 phase is greater than that in the G2 phase, and that the degree of the G1 phase dependency of the class I antigen density of leukemic T cell lines and lymphoblastoid B cell lines are significantly greater than that of activated T cells. The mean GffG~ ratios of the relative quantity of class II antigens on the surface of activated T cells and lymphoblastoid B cell lines were 1.86 ± 0.14 and 1.31 ± 0.05, respectively [Figure 5 (a)]. The former was significantly higher than the latter (p < 0.008). The mean G2/G, ratios of class II antigen density on the surface of activated T cells and lymphoblastoid B cell lines were 1.12 ± 0.06 and 0.79 - 0.02, respectively. The former was significantly higher than the latter (p < 0.001) [Figure 5 (b)]. T i m e Course of M H C A n t i g e n Expression on the Cell Surface A f t e r C h a n g i n g the C u l t u r e M e d i a In order to investigate the stability of M H C antigen quantity and density, determinations of the class I antigen expression on leukemic T cell lines were performed over 4 days. HPB-ALL and MOLT4 demonstrated no observable change in the percentage of cells in the GI and G~ phases, the quantity of antigen on the cell surface, the cell surface area, or the antigen density 1 and 2 days after seeding in fresh culture media (Table 2). Four days after seeding, the percentages of the cells in the G~ phase increased and both antigen quantity and cell surface area decreased in both G~ and G2 phases. The GffG1 ratios of the antigen density on both HPB-ALL and MOLT4 increased on day 4, in relation to the fact that the increase in the surface antigen densities on the cells in the G2 phase was greater than that in the G1 phase (Table 2). The cell numbers on days 0, 1, and 2 were 0.2, 0.3, and 0.5 (× 106/ml) for HPB-ALL, respectively, and 0.3, 0.5 and

FIGURE 5 The G2/Gt ratio of MHC class lI antigen expression on the surface of the cell lines. The means of the GffG] ratios of antigen quantity (#) and antigen density (##) in activated T cells (top) were significantly greater than that in lymphoblastoid B cell lines (bottom) (p < 0.008 and p < 0.001, respectively) (5a,5b). The difference in the mean GffGI, ratio of the antigen density on activated T cell lines between class I antigens (Figure 4b, **) and class II antigens (Figure 5b, ##) was significant (p < 0.008), while that on lymphoblastoid B cell lines between class I antigens (Figure 4b) and class II antigens (Figure 5b) was not significant.

5a

CELL LINE PHA-T1 "1[ C l o n e g "t[ C l o n e 18 ~[ C l o n e 2 3 1| Meen~SEM

1

-J I

:

i

4,

~

4,4,

It--1

EBV-B2

If~l

l| II

LIIz 0 0 7 Dludl

,

SIS I[ MeantSEM I ~ Quantity

•

I

n

EBV-B1

5b

0

~

J i

1.2

P < 0.008 I

I

1.6

2.0

I"

0

I

0.6

~

P < 0.001

I

0.8

I

1.0 1.2

G2/G~ Ratio of Class I ! Antigen

I

1.4

I

1.60en~y

294

Y. Matsui et al.

TABLE 2

Time course of quantity and density of surface class I antigens after seeding Cell number

Cell lines HPB- ALL Day Day Day MOLT4 Day Day Day

Antigen quantity

Surface area

Antigen density

G1

G2

G1

G2

G1

G2

Gl

G2

GJGI

1a 2a 4b

53 d 48 72

47 52 28

1.00 ~ 1.00 0.89

1.21 1.23 1.15

1.00 e 0.93 0.84

1.51 1.41 1.26

1.00 1.08 1.07

0.81 0.87 0.92

0.81 0.81 0.86

1b 2b 4~

67 71 83

33 29 17

1.00 e 1.00 0.98

1.42 1.42 1.31

1.00 ' 0.96 0.76

1.72 1.70 1.02

1.00 1.04 1.27

0.82 0.83 1.27

0.82 0.80 1.00

"Cells were suspended in fresh medium with 0.2 × 106 cells/ml on day 0, and maintained without further seeding or splitting. bCells were suspended in fresh medium with 0.3 × 106 cells/ml on day 0, and maintained without further seeding or splitting. 'Cells were suspended in fresh medium with 0.5 × 106 cells/ml on day 0, and maintained without further seeding or splitting. °Relative cell number (%). ~Values of the surface antigen quantity and the cell surface area in both cell lines are normalized to the values in the G1 phase on day 1.

0.8 ( × 106/ml) for MOLT4, respectively. The cell numbers for HPB-ALL seeded at 0.3 ( x 106/ml) and that for MOLT4 seeded at 0.5 ( x 106/ml), reached 2.2 and 3.9 ( x 106/ml), respectively in 4 days. This indicates that both the cell surface antigen quantity and the cell surface area demonstrate specific values at each phase of the cell cycle, i.e., that the cell cycle-related fluctuation pattern of antigen density is constant when the cells are in logarithmic growth. DISCUSSION The present study was undertaken in order to compare the expression of class I and class II MHC antigens on activated T cells with that on T and B lymphoblastoid cell lines. Experiments were performed using a dual-beam flow cytometer to study the cell surface antigen expression during the cell cycle. It has previously been observed by us [31-34] and by others [29,30,36,40] that the combination of immunofluorescence and DNA analysis with H0342 using dual-beam flow cytometry offers a more precise and direct method for relative quantification of surface antigens and DNA content of individual lymphocytes during the cell cycle than do measurements influenced by synchrony induction methods and cell fixation as used in previous studies of cell cycle dependency [8-12]. The relationship of antigen quantity and cell size to the cell cycle, as observed following synchrony induction methods, may not reflect the true cell cycle distribution observable by DNA content analysis [11]. Both synchrony induction methods and fixation of cells reduce cell size, thus possibly altering the distribution of surface antigen density during the cell cycle [15,41]. Based on this fact, light scatter has been used to determine surface area of cells during the cell cycle. In most previous reports, no correlated measurements of cell size and DNA content were performed [8-11]. Using a dual-beam flow cytometer, cell cycle dependency of MHC antigen density on the cell surface is found to be constant when the cells are in logarithmic growth. These studies confirm previous results [8-12] showing that class I and

Cell Cycle Dependency of MHC Antigen Expression

295

class II MHC antigens are cell cycle-dependent, that the antigen quantity in the G2 phase is greater than that in the G1 phase, and that the quantity of these antigens on B lymphoblastoid cell lines is greater than that on T lymphoblastoid cell lines throughout the cell cycle (i.e., in both GI and G2 phases) [42,43]. The calculated G2/G1 ratio of the surface areas of cultured cells in a previous study was approximately 1.6 [44], a value which agrees with the present study. MHC antigen expression during the cell cycle has also been studied in murine cells; however, no correlated measurement of cell size and DNA content was performed [40,44-51]. It has been reported that resting T cells (Go phase) demonstrate a rapid turnover of class I antigens, whereas activated T cells demonstrate a relatively decreased turnover [52]. It is also possible that G, and G2 cells demonstrate a different turnover of MHC antigens. Variation in surface molecules during the cell cycle has been accounted for by a limited period of gene transcription, a change in rate of synthesis, or a conformational change of molecules at the cell surface [53]. It has been reported that about 30-80 min after synthesis, mature MHC class I antigens are expressed at the cell surface of human lymphoblastoid B cell lines [54,55]. It has also been reported that a large pool of preformed determinants of MHC antigens is not present in human lymphocytes [56]. These findings suggest that the cell cycle dependency of the MHC antigen expression on the cell surface reflects regulation at the stage of translation, or earlier. We found that activated T cells demonstrate cell cycle dependency of MHC antigens, and that the quantity of class I antigens on activated T cells is lower than that on B lymphoblastoid cell lines but not different from T lymphoblastoid cell lines. These studies also demonstrate that the quantity of class II MHC antigens on activated T cells is not different from that on B lymphoblastoid cell lines. Additionally, in activated T cells, class I antigen density in the G2 phase was 11% less than in the G, phase, while class II antigen density in the G2 phase was 12% greater than that in the G, phase. Analyzing antigen quantity and surface area, it has been found that class I antigen density in the G2 phase was 17% less than that in the G1 phase in leukemic T cell lines, and that both class I and class II antigen densities in the G2 phase were 21% less than that in the G1 phase in lymphoblastoid B cell lines. These findings differ from previous results reporting that the density of MHC class I antigen remained constant during the cell cycle [9-11], or that the density of both class I and class II antigens increased in the G2 phase [8,11,12]. These discrepancies could result from differences in methods and/or in the number of lymphoblastoid cell lines used. Thus, we have studied 11 cell lines, whereas one or very few cell lines were used in the previous studies. It is possible that the difference in the cell cycle dependency of the MHC antigen expression between activated T cells and the leukemic or EBV-transformed cell lines is due to a difference in regulation of antigen synthesis during the transformed cell cycle. Alternatively, it is possible that the PHA-induced conditioned medium (PHACM) used in culturing activated T cells influences the cell cycle dependency of both class I and class II antigen expression on activated T cells. It has been reported that gamma interferon, often a component of PHA-CM, enhances the expression of both class I antigens [11,57,58] and class II antigens [59]. In this regard, we suggest that PHA-CM enhanced the expression of class I and class II antigens on PHA-activated T cells. In addition, our preliminary results demonstrate that PHA-CM increases the G2/G~ ratio of antigen quantity, suggesting that the increase of MHC antigens induced by PHA-CM occurs primarily in the G2 phase of activated T cells (data not shown). The lower density of MHC antigens on the leukemic cell lines in the G2 phase

296

Y. Matsui et al. compared to that in the G1 phase may allow a better chance for tumor cells in the G2 phase in vivo to escape from CTL-mediated lysis. It has been reported that the expression of a relatively large quantity of MHC antigens on the surface of the target tumor cells is required for efficient CTL-mediated lysis to occur [1]. In one study, neither H-2 antigen expression or CTL-mediated lysis was found to fluctuate with cell cycle progression in a murine tumor cell line [40]. In contrast, in a different system, a tumor target cell line was much more susceptible in the G1 phase, relative to other phases, to the cytotoxic effect of CTLs derived by immunization with MHC incompatible cells [60]. It is, however, unclear whether or not the differences in antigen expression demonstrated in the present study are large enough to have any meaningful biological impact in CTL-mediated lysis. More such studies are required in order to determine the significance or existence of selective CTL-mediated lysis of cells in different cell cycle phases. A number of studies have reported that cell functions, cellular components, and expression of surface molecules change during the cell cycle [61]. Production of immunoglobulins G and M is greatest during the late G1 and S phase [62]. Production and absorption of interleukin-2 occur during the late G, phase of stimulated mononuclear cells [63]. Macrophage-mediated antiproliferation activity to normal and neoplastic cells is dependent on the G~ and S phase of affected target cells [64]. Basophilic leukemia cells in the S/G2/M phase contain more histamine and release more of their histamine after activation than that in the G1 phase [65]. Colony-stimulating factor (CSF) acts on CSF-dependent cells in the G1 phase [66]. Exocytosis and endcytosis of transferrin are arrested during mitosis of Hela cells [67]. In contrast to the heterogeneous pattern of cell cycle dependency in the cell functions above, the quantity of some surface molecules is not dependent on the G, phase and maximum values are found in the S/GriM phase [53,68-73]. This may be a result of greater cell size in the S/GriM phase. This is shown for molecules, such as blood group antigens [53], a binding site of wheat-germ agglutinin [68], thymus leukemia antigen [69], melanoma-associated antigens [70], interleukin-2 receptor [71], transferrin receptor [72], Fc receptor [72], and receptor for B cell differentiation factor [73]. Recently, we have observed that the quantity of T cell differentiation antigens, CD3, CD4, and CD8, demonstrated maximum values at the S/G2/M phase of activated T cells. The density of the antigens, however, demonstrated maximum values at the G1 phase. Further, the GflG1 ratio of density of CD3, CD4, and CD8 antigens was significantly less than that of both class I and class II MHC antigens in T cell clones, but comparable to that of class I antigens in PHA-activated T cells (manuscript in preparation). The relationship of cell function to surface antigen density, which demonstrate heterogeneous patterns in the cell cycle, may be more important than cell function to surface antigen quantity. The most important observation of the present study is the finding of a greater density of class II antigens on activated T cells in the G2 phase, an observation which contrasts with that of B lymphoblastoid cell lines. It is interesting that such activated T cells can be distinguished from T and B lymphoblastoid cell lines by cell cycle studies. Because it is known that soluble factor(s), such as gamma interferon, can increase the MHC antigen density, it would be important to study the effect of PHA-CM and soluble factors on the density of MHC antigens in both T and B lymphoblastoid cell lines. Such studies should provide an understanding of the sequence of surface cellular events that could produce efficient cellular interactions between activated T cells and other cells during the immune response.

Cell Cycle Dependency of MHC Antigen Expression

297

ACKNOWLEDGMENTS We thank Dr. Ryohei Okamoto for critical review of the manuscript, Miss Lisa Christenson for excellent technical assistance in flow cytometry, Dr. Michael J. Sheehy for providing T cell clones, Dr. Deborah J. Anderson for helpful discussions, and Miss Elizabeth E. Eynon and Mr. Glen B. Glater for editorial assistance with the manuscript.

REFERENCES 1. Plata F, Tilkin AF, Levy JP, Lilly F: Quantitative variations in the expression of H-2 antigens on murine leukemia virus-induced tumor cells can affect the H-2-restriction patterns of tumor-specific cytolytic T lymphocytes. J Exp Med 154:1795, 1981. 2. Henry C, Chan EL, Kodlin D: Expression and function of I region products on immunocompetent cells. II. I region products in T-B interaction. J Immunol 119:744, 1977. 3. Bottomly K, Jones B, Kaye J, Jones III F: Subpopulations of B cells distinguished by cell surface expression of Ia antigens. Correlation of Ia and idiotype during activation by cloned Ia-restricted T cells. J Exp Med 158:265, 1983. 4. Janeway CA, Bottomly K, Babich J, Conrad P, Conzen S, Jones B, Kaye J, Katz M, McVay L, Murphy DB, Tire J: Quantitative variation in Ia antigen expression plays a central role in immune regulation. Immunol Today 5:99, 1984. 5. Matis LA, Jones PP, Murphy DB, Hedrick SM, Lerner EA, Janeway CA, McNicholas JM, Schwartz RH: Immune response gene function correlates with the expression of an Ia antigen. II. A quantitative deficiency in A~: E~ complex,expression causes a corresponding defect in antigen-presenting cell function. J Exp Med 155:508, 1982. 6. Matis LA, Glimcher LH, Paul WE, Schwartz RH: Magnitude of response of histocompatibility-restricted T-cell clones is a function of the product of the concentrations of antigen and Ia molecules. Proc Natl Acad Sci USA 80:6019, 1983. 7. Brown MF, Cook RG, Van M, Rich RR: Cloned human T cells synthesize Ia molecules and can function as antigen presenting cells. Hum Immunol 11:219, 1984. 8. Everson LK, Plocinik BA, Rogentine GN: HL-A expression on the G~, S and G2 cell-cycle stages of human lymphoid cells. J Nat Cancer Inst 53:913, 1974. 9. Pellegrino MA, Ferrone S, Cooper NR, Dierich MP, Reisfeld RA: Variation in susceptibility of a human lymphoid cell line to immune lysis during the cell cycle. Lack of correlation with antigen density and complement binding. J Exp Med 140:578, 1974. 10. Pellegrino MA, Ferrone S, Natali PG, Pellegrino A, Reisfeld RA: Expression of HL-A antigens in synchronized cultures of human lymphocytes. J Immunol 108:573, 1972. 11. Sarkar S, Glassy MC, Ferrone S,Jones OW: Cell cycle and the differential expression of HLA-A, B and HLA-DR antigens on human B lymphoid cells. Proc Natl Acad Sci USA 77:7297, 1980. 12. Killander D, Klein E, Levin A: Expression of membrane-bound IgM and HL-A antigens on lymphoblastoid cells in different stages of the cell cycle. Eur J Immunol 4:327, 1974. 13. Klevecz RR, Forrest GL: Metabolic regulation in the cell cycle. In: G H Rothblat, YJ Cristofalo, Eds., Growth, nutrition, and metabolism of cells in culture. New York, Academic Press, 1977, p. 149.

298

Y. Matsui et al. 14. Mitchison JM: Changing perspectives in the cell cycle. In: PCL John, Ed., The cell cycle. New York, Cambridge University Press, 1981, p. 1. 15. Braylan RC, Benson NA, Nourse V, Kruth HS: Correlated analysis of cellular DNA, membrane antigens and light scatter of human lymphoid cells. Cytometry 2:337, 1982. 16. Evans RL, FaldettaTJ, Humphreys RE, Pratt DM, Yunis EJ, Schlossman SF: Peripheral human T cells sensitized in mixed leukocyte culture synthesize and express la-like antigens. J Exp Med 148:1440, 1978. 17. Suciu-Foca N, Susinno E, McKiernan P, Rohowsky C, Weiner J, Rubinstein P: DRw determinants on human T cells primed against allogeneic lymphocytes. Transplant Proc 10:845, 1978. 18. Foley GE, Lazarus H, Farber S, Uzman BG, Boone BA, McCarthy RE: Continuous culture of human lymphoblasts from peripheral blood of a child with acute leukemia. Cancer 18:522, 1965. 19. Adams RA, Pothier L, Flowers A, Lazarus H, Farber S, Foley GE: The question of stemlines in human acute leukemia. Comparison of cells isolated in vitro and in vivo from a patient with acute lymphoblastic leukemia. Exp Cell Res 62:5, 1970. 20. Minowada J, Ohnuma T, Moore GE: Rosette-forming human lymphoid cell lines. I. Establishment and evidence for origin of thymus-derived lymphocytes. J Nat Cancer Inst 49:891, 1972. 21. MinowadaJ, Janossy G, Greaves MF, Tsubota T, Srivastava BIS, Morikawa S, Tatsumi E: Expression of an antigen associated with acute lymphoblastic leukemia in human leukemia-lymphoma cell lines. J Nat Cancer Inst 60:1269, 1978. 22. Schneider U, Schwenk HU, Bornkamm G: Characterization of EBV-genome negative "null" and "T" cell lines derived from children with acute lymphoblastic leukemia and leukemic transformed non-Hodgkin lymphoma. Int J Cancer 19:621, 1977. 23. Klein E, Klein G, Nadkarni JS, Nadkarni JJ, Wigzell H, Clifford P: Surface IgMkappa specificity on a Burkitt lymphoma cell in vivo and in derived culture lines. Cancer Res 28:1300, 1968. 24. Spiro RC, DeMartino JL, Boto W, Lazarus H, Humphreys RE: Comparison of membrane proteins of Burkitt's lymphoma and EBV-transformed B lymphoblast cell lines and of Con A-activated T lymphocytes and T lymphoblast cell lines. Leuk Res 3:315, 1979. 25. Collins SJ, Gallo RC, GaUagher RE: Continuous Growth and differentiation of human myleoid leukaemic cells in suspension culture. Nature 270:347, 1977. 26. Sheehy MJ, Yunis EJ, Agostini RM, Quintieri FB, Leung DYM, Geha RS, Yunis EJ: Morphology of human T lymphocyte clones. Lab Invest 48:549, 1983. 27. Barnstable CJ, Bodmer WF, Brown G, Galfre G, Milstein C, Williams AF, Ziegler A: Production of monoclonal antibodies to group A erythrocytes, HLA and other human cell surface antigens-new tools for genetic analysis. Cell 14:9, 1978. 28. Nadler LM, Stashenko P, Hardy R, PesandoJM, Yunis EJ, Schlossman SF: Monoclonal antibodies defining serologically distinct HLA-D/DR related Ia-like antigens in man. Hum Immunol 1:77, 1981. 29. Arndt-Jovin DJ, Jovin TM: Analysis and sorting of living cells according to deoxyribonucleic acid content. J Histochem Cytochem 25:585, 1977. 30. Loken MR: Simultaneous quantitation of Hoechst 33342 and immunofluorescence on viable cells using a fluorescence activated cell sorter. Cytometry 1:136, 1980.

Cell Cycle Dependency of MHC Antigen Expression

299

31. Shapiro HM: Flow cytometric estimation of DNA and RNA content in intact cells stained with Hoechst 33342 and pyronin Y. Cytometry 2:143, 1981. 32. Shapiro HM, Feinstein DM, Kirsch AS, Christenson L: Multistation multiparameter flow cytometry: Some influences of instrumental factors on system performance. Cytometry 4:11, 1983. 33. Williams JM, Shapiro HM, Milford EL, Strom TB: Multiparameter flow cytometric analysis of lymphocyte subpopulation activation in lectin-stimulated cultures. J Immunol 128:2676, 1982. 34. Comer T, Williams JM, Christenson L, Shapiro HM, Strom TB, Strominger J: Simultaneous flow cytometric analysis of human T cell activation antigen expression and DNA content. J Exp Med 157:461, 1983. 35. Loken MR, Herzenberg LA: Analysis of cell populations with a fluorescence-activated cell sorter. Ann NY Acad Sci 254:163, 1975. 36. Mullaney PF, Crowell JM, Salzman GC, Martin JC, Hiebert RD, Goad CA: Pulseheight light-scatter distributions using flow-systems instrumentation. J Histochem Cytochem 24:298, 1976. 37. Jones EA, Goodfellow PN, Bodmer JG, Bodmer WF: Serological identification of HL-A-linked human 'Ia-type' antigens. Nature 256:650, 1975. 38. Martin PJ, Giblett ER, HansenJA: Phenotyping human leukemic T-cell lines: Enzyme markers, surface antigens, and cytogenetics. Immunogenetics 15:385, 1982. 39. Boss MA, Delia D, Robinson JB, Greaves MF: Differentiation-linked expression of cell surface markers on HL-60 leukemic cells. Blood 56:910, 1980. 40. Harris DT, Sekaly RP: Target cell susceptibility to immune lysis and expression of MHC antigens are independent of position in the cell cycle. J Immunol 133:40, 1984. 41. Lerner RA, Hodge LD: Gene expression in synchronized lymphocytes. Studies on the control of synthesis of immunoglobulin polypeptides. J Cell Physio177:265,1971. 42. Trucco M, dePetris S, Garotta G, Ceppellini R: Quantitative analysis of cell surface HLA structures by means of monoclonal antibodies. Hum Immunol 1:233, 1980. 43. Pellegrino MA, Belvedere M, Pellegrino AG, Ferrone S: B peripheral lymphocytes express more HLA antigens than T peripheral lymphocytes. Transplantation 25:93, 1978. 44. Graham JM, Sumner MCB, Curtis DH, Pasternak CA: Sequence of events in plasma membrane assembly during the cell cycle. Nature 246:291, 1973. 45. Cikes M: Relationship between growth rate, cell volume, cell cycle kinetics, and antigenic properties of cultured murine lymphoma cells. J Nat Cancer Inst 45:979, 1970. 46. Cikes M, Friberg S: Expression of H-2 and Moloney leukemia virus-determined cellsurface antigens in synchronized cultures of a mouse cell line. Proc Natl Acad Sci USA 68:566, 1971. 47. Pasternak CA, Warmsley AMH, Thomas DB: Structural alterations in the surface membrane during the cell cycle. J Cell Biol 50:562, 1971. 48. Cikes M, Friberg S, Klein G: Quantitative studies of antigen expression in cultured murine lymphoma cells. II. Cell-surface antigens in synchronized cultures. J Nat Cancer Inst 49:1607, 1972. 49. Sumner MCB, Collin RCLS, Pasternak CA: Synthesis and expression of surface antigens during the cell cycle. Tissue Antigens 3:477, 1973. 50. Lanier LL, Warner NL: Cell cycle related heterogeneity of Ia antigen expression on

300

Y. Matsui et al. a murine B lymphoma cell line: Analysis by flow cytometry. J Immunol 126:626, 1981. 51. Monroe JG, Cambier JC: Level of mIa expression on mitogen-stimulated murine B lymphocytes is dependent on position in cell cycle. J Immunol 130:626, 1983. 52. Monos DS, Cooper HL: Rapid turnover of HLA proteins in quiescent lymphocytes: Proposed connection with immunologic surveillance. J Immunol 131:341, 1983. 53. Thomas DB: Cyclic expression of blood group determinants in murine cells and their relationship to growth control. Nature 233:317, 1971. 54. Krangel MS, Orr HT, Strominger JL: Assembly and maturation of HLA-A and HLA-B antigens in vivo. Cell 18:979, 1979. 55. Owen MJ, Kissonerghis AM, Lodish HF: Biosynthesis of HLA-A and HLA-B antigens in vivo. J Biol Chem 255:9678, 1980. 56. Turner MJ, Strominger JL, Sanderson AR: Enzymic removal and re-expression of a histocompatibility antigen, HL-A 2, at the surface of human peripheral lymphocytes. Pro Natl Acad Sci USA 69:200, 1972. 57. Lindahl P, Leary P, Gresser I: Enhancement of the expression of histocompatibility antigens of mouse lymphoid cells by interferon in vitro. Eur J Immunol 4:779, 1974. 58. Heron I, Hokland M, Berg K: Enhanced expression of 32-microglobulin and HLA antigens on human lymphoid cells by interferon. Pro Natl Acad Sci USA 75:6215, 1978. 59. Wong GH, Clark-Lewis I, McKimm-Breschkin JL, Harris AW, Schrader JW: Interferon-~/induces enhanced expression of Ia and H-2 antigens on B lymphoid, macrophage, and myeloid cell lines. J Immunol 131:788, 1983. 60. Leneva NY, Svet-Moldavsky GJ: Susceptibility of tumor cells in different phases of the mitotic cycle to the effect of immune lymphocytes. J Nat Cancer Inst 52:699, 1974. 61. Lloyd D, Poole RK, Edwards SW: Biosynthesis of macromolecular components: RNA, proteins, and other cellular constituents. In: The cell division cycle. New York, Academic Press, 1982, p.143. 62. Buell DN, Fahey JL: Limited periods of gene expression in immunoglobulin-synthesizing cells. Science 164:1524, 1969. 63. Stadler BM, Dougherty SF, Farrar JJ, Oppenheim JJ: Relationship of cell cycle to recovery of IL 2 activity from human mononuclear cells, human and mouse T cell lines. J Immunol 127:1936, 1981. 64. Hamilton TA, Fishman M, Crawford G, Thomas Look A: Sensitivity to macrophagemediated cytostasis is cell cycle dependent. Cell Immunol 69:363, 1982. 65. Meyer C, Wahl LM, Stadler BM, Siraganian RP: Cell cycle associated changes in histamine release from rat basophilic leukemia cells separated by counterflow centrifugal elutriation. J Immunol 131:911, 1983. 66. Pluznik DH, Cunningham RE, Noguchi PD: Colony-stimulating factor (CSF) controls proliferation of CSF-dependent cells by acting during the G1 phase of the cell cycle. Proc Natl Acad Sci USA 81:7451, 1984. 67. Sager PR, Brown PA, Berlin RD: Analysis of transferrin recycling in mitotic and interphase Hela Cells by quantitative fluorescence microscopy. Cell 39:275, 1984. 68. Fox TO, Sheppard JR, Burger MM: Cyclic membrane changes in animal cells: Transformed cells permanently display a surface architecture detected in normal cells only during mitosis. Proc Natl Acad Sci USA 68:244, 1971.

Cell Cycle Dependency of MHC Antigen Expression

301

69. Williams JL, Pickering JW, Wolcott M: Selective cell surface expression of thymus leukemia antigen during S phase of the cell cycle. J Immunol 122:1121, 1979. 70. Burchiel SW, Martin JC, Imai K, Ferrone S, Warner NL: Heterogeneity of HLA-A, B,Ia-like, and melanoma-associated antigen expression by human melanoma cell lines analyzed with monoclonal antibodies and flow cytometry. Cancer Res 42:4110, 1982. 71. Williams JM, Loertscher R, Comer T, Reddish M, Shapiro HM, Carpenter CB, Strominger JL, Strom TB: Dual parameter flow cytometric analysis of DNA content, activation antigen expression, and T cell subset proliferation in the human mixed lymphocyte reaction. J Immunol 132:2330, 1984. 72. London L, Perussia B, Trinchieri G: Induction of proliferation in vitro of resting human natural killer cells: Expression of surface activation antigens. J Immunol 134:718, 1985. 73. Brooks KH, Uhr JW, Vitetta ES: Cell cycle-related expression of the receptor for a B cell differentiation factor. J Immunol 134:742, 1985.