The estrogen-responsive 110K and 74K rat uterine secretory proteins are structurally related to complement component C3

Vol. 158, No. 3, 1989 February 15, 1989

THE

BIOCHEMICAL

ESTROGEN-RESPONSIVE

PROTEINS

ARE

STRUCTURALLY

1lOK

AND BIOPHYSICAL RESEARCH COMMUNICATIONS Pages 898-905

AND

RELATED

74K RAT

UTERINE

TO COMPLEMENT

SECRETORY COMPONENT

C3

Paola C. Kuivanen*, RhodessaB. Capulong*, Richard N. Harkins+ and Eugene R.DeSombrea with the technical assistanceof Barbara E. Nowakowski* *The Ben May Institute, The University of Chicago, 5841 South Maryland Avenue Chicago, IL 60637 +Triton Biosciences Inc., 1501 Harbor Bay Parkway, Alameda, CA 94501 Received December 27, 1988 Estrogens stimulate the synthesis of specific secretory proteins in the rat uterus. Here we show that two of these, polypeptides of relative molecular weight 110,000 (1lOK) and 74,000 (74K), are structurally related to C3, the third component of complement, a glycoprotein that plays a central role in regulating complementmediated inflammatory and immune responses. The similarities were based on the observations that (1) NH2-terminal amino acid sequence of the 74K polypeptide showed sequence homology with the 8 chain of mouse C3, (2) comparison of the electrophoretic mobilities of the 1lOK and 74K polypeptides in the presence and absence of reducing agents revealed that they were disulfide-linked subunits of a protein of Mr -180,000, (3) the native protein was immunoreactive with antibodies specific for rat C3, and (4) both polypeptides were immunoprecipitated with Press,Inc. antibodies to rat C3. Q1989Academic SUMMARY.

178-Estradiol

(E2) stimulates a myriad of responsesin the uterus that result in

biochemical and morphological changes, culminating in uterine growth and functional differentiation (l-3). In an effort to understand the molecular basis of the mechanism of estrogen action in the uterus, several groups studied the effect of estrogens on uterine secretory estrogen-stimulated proteins (4-9).

protein synthesis and identified We reported that E2 differentially

a number of modulates the

synthesis of secretory proteins in the rat uterus, dramatically inhibiting the synthesis of some, while stimulating the synthesis of others in a sequentially ordered manner (10). We identified two polypeptides, relative molecular mass (Mr) 110,000 (1lOK) and 74,000 (74K), among the proteins whose in vitro synthesis is precociously 1To whom correspondence should be addressed. used are BSA bovine serum albumin, CAPS 3- c clohexylam$; I-propanesulfonic acid, DTT dithiothreitol, E 178-estradiol, &T-free lil metliionine-free minimum essential medium, ii relativep~$e~lar mass, N C M nitrocellulose membrane, PTH phenylthiohydatntoin, polyvinylrdene difluoride. SD S- PA G E , sodium dodecyl sulfate-denaturing polyacrylamide electrophoresis, TBS-Tween 50 mM Tris pH 7.4, 150 mM NaCl and 0.2% Tween 20, T E ?! trichloroacetic acid. ABBREVIATIONS

ooo6-291x/89$1.50 Copyright All rights

8 1989 by Acaakmic Press, Inc. of reproduction in any form reserved.

898

Vol. 158, No. 3, 1989

BIOCHEMICAL

AND BtOPHYSlCAL

RESEARCH COMMUNICATIONS

stimulated in uteri of immature rats treated with E2 and whose increased synthesis is inhibited by progestins (11).

We observed that these two polypeptides are not

synthesized in uteri from adult ovariectomized rats but are synthesized following exogenous administration of E2. Synthesis of the two polypeptides is evident during proestrus and estrus but not during metestrus and diestrus (12). Lyttle and coworkers observed similar hormonal regulation of two secretory proteins, Mr 1 15,000 and 65,000 (5-8). Though their hormonal regulation is established, the function of these polypeptides in the E2-stimulated uterus is not known. We now report that we have identified structural similarities between the 1lOK and 74K secretory proteins and C3, the third component of complement, a glycoprotein that plays a pivotal roIe in complement-mediated inflammatory MATERIALS

and immunological reactions. AND METHODS

. vitro labof uterine secretorv orotu. 22-day old (immature) Sprague-Dawley derived rats (Sasco Inc., Omaha, NE) were administered daily S.C. injections of 1 vg E2. Six hours after the third injection the animals were sacrificed and uteri were excised, slit open and incubated jn vitro at 37’C for 3 h with 100p.Ci L-[35S]methionine (sp. act. >lOOO Ci/mmol, Amersham Corp., Arlington Heights, IL) per ml MET-free MEM (Flow Laboratories, McLean, VA) as detailed previously (10). The secretory proteins were TCA-precipitated from the media and prepared for SDSPAGE. One-dimensional analytical SDS-PAGE was carried out using the discontinuous buffer system of Laemmli (13) as detailed elsewhere (10). TCA-precipitated protein was heated (9O’C for 4 min) in SDS-PAGE samplebuffer (50 mM Tris pH 6.8, 2% SDS, 10% sucroseand 0.005% bromophenolblue) in the presence of 5% DTT (DTT-reduced) or in its absence (non-reduced). Silver staining, autoradiography and fluorography were as described earlier (10). . n of 1lOK and 74K oolvneot&&. Luminal fluid was collected from mature ovariectomized rats three days after implantation of 0.5 mg E2 pellets (Innovative Research of America, Toledo, OH) and electrophoresed on onedimensional preparative 15% acrylamide SDS-PAGE gels, which were stained with 0.5% Coomassie Brilliant Blue R-250 and rapidly destained. The stained bands corresponding to the 1lOK and 74K polypeptides were cut from the gel and the protein electro-eluted (14). Each eluted protein preparation migrated as a single band when analyzed by SDS-PAGE with silver staining and by autoradiography of the radioiodinated products. . 100-200 pmol purified 1lOK and 74K polypeptides were separated on 10% acrylamide SDS-PAGE gels, electroblotted onto Immobilon PVDF membrane (Millipore Corporation, Bedford, MA) using 10 mM CAPS pH 11.0 without methanol as the transfer buffer (15, 16), stained with 0.1% CoomassieBlue in 50% methanol and destained. NH2-terminal sequence analysis was performed on an Applied Biosystems Model 475A gas phase sequencer (ABI, Foster City, CA) equipped with on-line HPLC detection of the PTH-derived amino acids. Coomassie stained protein bands were cut out of the PVDF membrane and placed directly onto filters precycled with BioBrene (ABI). . . Male New Zealand white rabbits were Production of pobhm.ldWlbodleS. injected with 50 pg purified 1lOK and 74K proteins, either in the acrylamide gel slice or as the electro-eluted protein, in Freund’s complete adjuvant (Difco Laboratories,. Detroit, MI) and boosted with 25 pg purified antigen in Freund’s incomplete adjuvant. The antisera C9 (antibodies to the 74K protein) and B19 (antibodies to the 1lOK protein) were of high titer and specificity. 60 pg aliquots of media and rat serum were separated on 7.5% gels and qualitatively electro-blotted onto NCM (17). Western blots were blocked with 3% BSA in TBS-Tween (1 h at room temperature) and probed with C9 (1:5000) and BI 9 (1:7500) followed by incubation with Protein A (Sigma Chemical Co., St. Louis, MO) labeled with Na[1251] by chloramine-T-catalyzed 899

Vol. 158, No. 3, 1989

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

iodination and autoradiography. Immunoblotting with sheep anti-rat C3 (1:lOOO) (ICN ImmunoBiologicals, Lisle, IL) was modified to substitute blocking with 3% ovalbumin and binding of [1251]rec-Protein G (Zymed Laboratories, Inc., South San Francisco, CA) to . the. bound sheep antibodies. . ltatlou. Aliquots of media proteins labeled with [35S] methionine were resuspended in RIPA buffer (50 mM Tris pH 7.2, 150 mM NaCl, 1% Triton X100, 1% deoxycholate and 0.1% SDS), precleared with normal sheep serum and Protein A-Sepharose CL-4B (Pharmacia LKB Biotechnology Inc., Piscataway, NJ), incubated with 3.5 pl sheep anti-rat C3 antibodies (overnight at 4°C). and precipitated by the addition of Protein A-Sepharose (6 h at 4’C). After centrifugation the pellet was washed and the antigen-antibody complexes eluted in 50 ~1 2 X SDS-PAGE sample buffer containing 10% DTT (lOO°C for 5 min).

RESULTS We observed

that

that co-migrated uterine secretory (Figure as much

luminal

fluid

electrophoretically proteins that were

1, Lanes 2 and 3). as 7%

from

and

which to purify mature rats

purified

to

homogeneity

elution.

The

sequence

analysis.

purified

the 1lOK

of the total

these polypeptides. implanted with E2

by

high-resolution

polypeptides The

fluid

respectively,

source from ovariectomized

immature

were

polyclonal

rats

contained

proteins

with the estrogen-responsive 110K and labeled in vitrq by uteri from E2-treated

In luminal

17%,

E2-treated

and 74K protein, Luminal pellets

preparative

used for

antibodies

antibody

Cg and B19

C

proteins making

74K rats

corresponded this

fluid was collected from and the two polypeptides SDS-PAGE production produced

and

electro-

and amino in male

acid rabbits

,110 K

-74K

1

2

3

4

56

Figure 1. Lanes 1 and 2. Autoradiographic image of the electrophoretic separation of media proteins labeled by&~ by uteri from control (Lane 1) and E2-treated rats . . equivalent . (Lane 2). Samples contained amounts of TCA-precipitated radioactivity, were reduced with P-mercaptoethanol and analyzed by 15% acrylamide SDS-PAGE. Lanes 3-6. Uterine luminal fluid from E2-treated rats analyzed by 15% acrylamide SDS-PAGE and silver stained (Lane 3) or electroblotted onto NCM and immunoblotted with rabbit preimmune serum (Lane 4) and antisera Cg (Lane 5) and Bl9 (Lane 6). 900

to

a reasonable

Vol. 158, No. 3, 1989

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

R.74K M-p

Chain

of C3

45

50

55

Figure 2. Comparison of the N-terminal amino acid sequence of the rat 74K protein with the deduced amino acid sequence of the amino terminus of the h chain of mouse C3 (19). Identical amino acids am boxed. R = rat. M = mouse. The numbering of the mouse C3 g chain corresponds to the sequence of the mouse preprocomplement component C3.

were

of

high

(Figure

titer

1, Lanes

polypeptides base

5 and 6).

revealed

C3, sharing

was

less

definitive;

than

used

for

-180,000

(20)

most

prominently

Lane

4).

is a disulfide-linked of

Analysis from

labeled

rats

protein

that

was

conditions

with

and 65K

by estogens, more

than

two

(Figure

labeled

by uteri

a number and

To

of

SDS-PAGE determine

conditions immunoblots

are of

in shown

unambiguously the

to media

non-reducing

the

the electrophoretic only some of

proteins

3A,

Lane and

3A,

absent were

Lanes

in

the

2 and uterine that

to be stimulated

however, rat

migration which were

not

concluded

observed We,

(7).

and

5). 74K

had resolved

uterus

(10)

and

of proteins under resolved by 7.5%

here. which

1lOK

proteins

3A,

of an estrogen-regulated

that they

protein

(Figure

changes

conditions

proteins

180K

b the

1lOK

(Figure

of

conditions

conspicuously

These

animals

compared

synthesized

(Figure the

whether

absence

-180,000

proteins

4).

presence

we

and

proteins

were

5),

control

secretory

conditions,

analysis

related

Lane

the

under of the

changes

non-reducing

3A, from

reported

estrogen-modulated

observed

Lane

determine

protein,

conditions

of Mr

of an cc chain

presence

proteins

3A,

the major

the subunits

reducing acrylamide

Figure

conditions

reducing

protein

to obtain

precursor

To

of M,

protein

more

non-reducing

as a doublet labeled

with

composed

media

data with

of the 8 chain

chain

(21).

the

under

protein

inadequate

of a larger in

that

under

major

2 and

polypeptides,

were

subunit

revealed

absent

also 3). Recently, Lyttle a& secretory protein of Mr 180,000 the 115K

bond

polypeptide

migrated

the

1, Lane proteins

a structure

of [35S]methionine-labeled

E2-treated

non-reducing

this

even was

as a single

disulfide

purified

data for the 1lOK

the signal

to form

the

fluid

homology

terminus

to be N-blocked,

is synthesized

of

sequence

The sequencing

protein,

luminal

of the NBRF

had

acids of the amino

not

74K

that

mobility

a protein

observed

appeared

of the

in

sequence

A search

2).

by a single

(Figure

under

it

antigens

acid

polypeptide

19) (Figure

that are linked

reducing

proteins

74K

processed

agents. by uteri

Under

(18,

polypeptide

respective amino

17 amino

glycoprotein

electrophoretic

m

rat

and subsequently

and a 8 chain the rat 74K

the

analysis

conclusive data. C3 is a serum

their

by microsequencing.

though

the

for

NH2-terminal

16 of the first

C3 (Mr 75,000)

reducing

The

that

of mouse

the

monospecific

was determined

library

mouse

and

and from 901

of

the

proteins

74K polypeptides, cultures of uteri

under we from

non-reducing

did a series of E2-treated rats,

Vol. 158, No. 3, 1989

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

200K-

.llOK

-llOK

IOOK92.5K-

- 74K .74K 69K-

I2

I2345

3456

B, A Figure 3. A. Fluorographic image of media proteins synthesized i.nx&~ by uteri from control (Lanes 2 and 3) and ES-treated rats (Lanes 4 and 5) analyzed by 7.5% acrylamide SDS-PAGE under DTT-reducing (Lanes 3 and 5) and non-reducing conditions (Lanes 2 and 4). Lane 1 is [14C]methylated protein standards: myosin (Mr 200,000), phosphorylase b (bands at Mr 100,000 and 92,500). and BSA (Mr 69,000). B. Western blots of 60 ug of protein secreted ind by uteri from E2-treated rats and analyzed by 7.5% acrylarnide SDS-PAGE under non-reducing (Lanes 1, 3 and 5) and DTT-reducing conditions (Lanes 2 and 4) were probed with C9 (Lanes 1 and 2). B 19 (Lanes 3 and 4). and sheep anti-rat C3 (Lane 5). The radiogram of the corresponding [35S]methionine-labeled samples is shown in A, Lanes 4 and 5, respectively. Lane 6 is a Western blot of 60 ug of serum from E2-treated rats, analyzed under non-reducing conditions, that was immmunoblotted with sheep antirat C3.

specifically

probing

proteins

for

analyzed

for the secreted (Figure

3B,

74K

Lane

However, antisera

of Mr -180,000

3B,

antisera

Sheep Lane

results

6). also

suggested

of a protein noted that serum

(Figure were

anti-rat

when

revealed

media

proteins

their

substrates

with

this

1lOK

74K

polypeptides

and

doublet

of M, -180,000 that was antigenically the native -180K uterine secretory

C3 (Figure

3B, Lanes

was

were

analyzed

a protein

that

These data suggested

was monospecific

specifically

the

Cg

for the 1lOK

with

1 and 3).

for

which

that

2) and Blg

specifically

3B, Lanes

Immunoblots

rat C3 proteins.

3B, Lane

reacted

specific

C3 serum,

reacted that

and

conditions

(Figure

both

a doublet

and Bl9

IIOK

4).

conditions

Cg

74K,

DTT-reducing

polypeptide

reducing

forms.

the

under

in

their

(Figure

under

3B,

Lane

was migrated

non-

migrated

as

that both the and 5).

disulfide-linked

related

protein

5 and 6) in that the latter

polypeptide

for C3 in rat serum were

closely

native

of media monospecific

reduced (Figure These subunits

to C3.

It should

different

from

as a single

be

native peptide

species. Polyclonal

antibodies

[35S]methionine-labeled immunoprecipitated 40,000

to

secreted a small

rat 1lOK

amount

C3 and of

specifically 74K

proteins

radioactive

regions. 902

immunoprecipitated (Figure

material

in

4)

and,

the

Mr

the in

addition,

66,000

and

Vol. 158, No. 3, 1989

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

-1lOK

1

2

Figore 4. Secretory proteins labeled ia y&~ by uteri from E2-treated rats before (Lane 1) and after (Lane 2) immunoprecipitation with sheep anti-rat C3. The immunoprecipitates were eluted with SDS-PAGE sample buffer containing reducing agent and analyzed by 15% acrylamide SDS-PAGE and fluorography. The secretory proteins (Lane 1) were analyzed as described in legend to Figure 1.

DISCUSSION The

close

relationship

and C3 was revealed chain

of

mouse

non-reducing labeled

antigens

for

1lOK

also

revealed

the amino C3a (M, C3

C3,

for

is

tentative

sequence

When

activated

form

superfamily acid

similarities, homology polypeptides

of of

to rat C3.

C3 (Mr cleaved

in

the

and with

both

C3 is

permeability reacts

covalently

viral

infectivity

complement similar

structural C3

cleaved,

and with (26).

C3

all

share

functions. similarities that

a

which and

the may

903

1lOK be

of

of

protein

homologous

of

complement in

for

example,

C3,

that,

domain, amino

74K

for

which

the

proteins of

acids

is responsible

effecting,

and

data

in that it

of C3.

C3b,

component

distinct

data

and rat anaphylatoxin

25),

includes

of the

16 amino

C3a,

identification

of they

one

and

antisera

sequence

this

pathways

complexes,

which

The

(22)

alternate

(24,

is

with

acid

with

112,933)

releasing

(27),

amino

from the c( chain

and

immune

and the p

reducing

protein

13 of the first

chemotaxis

system

homology,

suggests

classical

proteins

and the immunoprecipitation

sharing

of mouse

under

native

was consistent

selectively

sequence and

C3,

secretory

polypeptide

mobilities the

NH2-terminal

it was,

with

uterine

of the 74K

of

a peptide

activated,

the

homology

74K

electrophoretic

though

homology

protein

and

and C3 proteins,

of the Q chain (23),

vascular

neutralization

74K

antibodies

a key

1lOK

immunoreactivity

with

termini

rat

differential the

protein,

activation. increased

amino

their

the llOK,

-9,000)

the

here by the sequence

conditions,

monospecific the

between

acid uterine

components

its

C4,

C5

based

on

structural sequence secretory of

this

Vol. 158, No. 3, 1989

structural

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

superfamily and may function

in the hormonal regulation of vascular

permeability and the immune system of the rat uterus. Our earlier studies showed that the in vitro synthesis of the 1lOK and 74K secretory polypeptides was increased greater than 60-fold in uteri from E2-treated immature rats (10).

More recently, immunocytochemical studies have localized the

increased accumulation of these antigens in the apical portion of the cytoplasm of luminal and glandular epithelia of uteri from E2-treated rats (L. Puy, P. Kuivanen, and E. DeSombre, manuscript in preparation). Together the data are consistent with estrogen regulating the expression of the 1lOK and 74K polypeptides in the rat uterus. Estrogens have marked stimulatory effects on several aspects of the secretory immune system in the rat uterus, increasing the accumulation of IgA, IgG and secretory component in uterine secretions (28) and increasing the synthesis of the IgM

class of antibodies (29).

However, estrogen regulation of C3, or other

components of that superfamily of the complement system, has not been previously reported in the rat uterus. Hormonal control of C3 has been suggested in other systems. Hydrocortisone has been observed to specifically increase the synthesis of C3 several fold in a rat hepatoma cell line (30). In humans, C3 concentrations are significantly elevated in maternal serum at the time of parturition (31). The available information indicates, but does not prove, that the 1lOK and 74K polypeptides are synthesized in uterine epithelial cells. Major sites of synthesis of C3 are liver cells (32) and, to a lesser extent, monocytes, macrophages, splenic and lymphoid cells, and other tissues of the reticuloendothelial system.

Recently, it was

reported that C3 is synthesized and secreted by human fibroblasts (33) and capillary endothelial cells (34, 35) in primary culture. The site of synthesis of the 1lOK and 74K polypeptides in the E2-stimulated uterus is an important factor in understanding their function and for studying the mechanism of hormonal regulation of their expression. Based on the results presented here, it should be possible to prepare specific cDNA probes for these markers and localize the mRNAs for the IlOK and 74K polypeptides to show the cellular origin of their estrogen-dependent synthesis. ACKNOWLEDGMENTS. We thank Paul Gilna for assistancein computer searchesof the Protein Databases and Alun Hughes for iodination of proteins. This work was supportedin part by NC1 Grant CA 09183 and NIH Grant HD 15513 to E.R.D. R.B.C. was a Dorothy M. and Hugh A. Edmondson-Beatrice Garber Summer Research Scholar supported by the Richter Fund for Undergraduate Research. REFERENCES 1. Szego, C.M., and Roberts, S. (1953) Recent Prog. Horm. Res. 8, 419-469. 2. Clark, J.H., Peck, E.J. Jr., Hardin, J.W., and Eriksson, H. (1978) In Receptors and Hormone Action (B.W. O’Malley and L. Birnbaumer, Eds.), Vol. 2, pp. 1-31. Academic Press, New York. 3. Katzenellenbogan, B.S., Bhakoo, H.S., Ferguson, E.R., Lan, N.C., Tatee, T., Tsai, TL.S., and Katzenellenbogen, J.A. (1979) Recent Prog. Horm. Res. 35. 259-300. 4. Surani, M.A.H. (1977) J. Reprod. Fert. 50, 289-296. 5. Komm, B.S., Keeping, H.S., Sabogal, G., and Lyttle, C.R. (1985) Biol. Reprod. 32, 443450. 904

Vol. 158, No. 3, 1989

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

6. Komm, B.S., Rusling, D.J., and Lyttle, C. R. (1986) Endocrinology 118, 2411-2416. 7. Lyttle, C.R., Wheeler, C,, and Komm, B.S. (1987) In Cell and Molecular Biology of the Uterus (W.W. Leavitt, Ed.), pp. 119-136. Plenum Publishing Corporation, New York. 8. Wheeler, C., Komm, B.S., and Lyttle, C.R. (1987) Endocrinology 120, 919-923. 9. Takeda, A., Takahashi, N., and Shimizu, S. (1988) Endocrinology 122, 105113. 10. Kuivanen, P.C., and DeSombre, E.R. (1985) J. Steroid Biochem. 22, 439-451. 11. DeSombre, E.R., and Kuivanen, P.C. (1985) Seminars in Oncology 12, 6-11. 12. Kuivanen, P.C., and DeSombre, E.R. (1985) In Program of the 67th Annual Meeting of The Endocrine Society, p. 51 (Abstract). 13. Laemmli, U.K. (1970) Nature (London) 227, 680-685. 14. Hunkapiller, M.W., Lujan, E., Ostrander, F., and Hood, L.E. (1983) Methods Enzymol. 91, 227-236. 15. Matsudaira, P. (1987) J. Biol. Chem. 262, 10035-10038. 16. LeGendre, N., and Matsudaira, P. (1988) BioTechniques 6, 154-159. 17. Towbin, H., Staehelin, T., and Gordon, J. (1979) Proc. Natl. Acad. Sci. U.S.A. 76, 43504354. 18. Gyongyossy, M.I.C., and Assimeh, S.N. (1977) J. Immunol. 118, 1032-1035. 19. Lundwall, A., Wetsel, R.A., Domdey, H., Tack, B.F., and Fey, G.H. (1984) J. Biol. Chem. 259, 13851-13856. 20. Brade, V., Hall, R.E., and Colten, H.R. (1977) J. Exp. Med. 146, 759-765. 21. Weiler, J.M., and Hobbs, M.V. (1987) In Autoimmunoregulation and Autoimmune Disease (J.M. Cruse and R.E. Lewis, Jr., Eds.), Concepts Immunopathol., Vol. 4, pp. 103-128. Karger, Basel, Switzerland. 22. Wetsel, R.A., Lundwall, A., Davidson, F., Gibson, T., Tack, B.F., and Fey, G.H. (1984) J. Biol. Chem. 259, 13857-13862. 23. Jacobs, J.W., Rubin, J.S., Hugli, T.E., Bogardt, R.A., Mariz, I.K., Daniels, J.S., Daughaday, W.H., and Bradshaw, R.A. (1978) Biochemistry 17, 5031-5038. 24. Hugli, T.E. (1975) J. Biol. Chem. 250, 8293-8301. 25. Hugli, T.E. (1981) CRC Crit. Rev. Immunol. 1, 321-366. 26. McNearney, T.A., Odell, C., Holers, V.M., Spear, P.G., and Atkinson, J.P. (1987) J. Exp. Med. 166, 1525-1535. 27. Bentley, D.R. (1988) In Genetics of Complement (M. Adinolfi, Ed.), Expl. Clin. Immunogenet., Vol. 5, pp. 69-80. Karger, Basel, Switzerland. 28. Wira, C.R., Sullivan, D.A., and Sandoe,C.P. (1983) Ann. N. Y. Acad. Sci. 409, 534-551. 29. Myers, M.J., and Petersen, B.H. (1985) Int. J. Immunophatmac. 7, 207-213. 30. Strunk, R.C., Tashjian, A.H. Jr., and Colten, H.R. (1975). J. Immunol. 114, 331-335. 31. Propp, R.P., and Alper, CA. (1968) Science 162, 672-673. 32. Alper, C.A., Johnson, A.M., Birtch, A.G., and Moore, F.D. (1969) Science 163, 286288. 33. Whitehead, A.S., Solomon, E., Chambers, S., Bodmer, W.F., Povey, S., and Fey, G. (1982) Proc. Natl. Acad. Sci. U.S.A. 79, 5021-5025. 34. Ueki, A., Sai, T., Oka, H., Tabata, M., Hosokawa, K., and Mochizuki, Y. (1987) Immunology 61, 11-14. 35. Warren, H.B., Pantazis, P., and Davies, P.F. (1987) Am. J. Pathol. 129, 9-13.

905