Citation
The immunobiology of graft versus host disease and its attempted prevention using naturally occurring suppressor factors

Material Information

Title:
The immunobiology of graft versus host disease and its attempted prevention using naturally occurring suppressor factors
Creator:
Jadus, Martin Robert, 1953-
Publication Date:
Language:
English
Physical Description:
xvi, 244 leaves : ill., graphs ; 29 cm.

Subjects

Subjects / Keywords:
Animals ( jstor )
Antigens ( jstor )
In vitro fertilization ( jstor )
Liver ( jstor )
Lymphocytes ( jstor )
Mortality ( jstor )
Neonates ( jstor )
Reactivity ( jstor )
Spleen ( jstor )
Splenocytes ( jstor )
Dissertations, Academic -- Pathology -- UF ( mesh )
Graft vs Host Reaction ( mesh )
Pathology thesis Ph.D ( mesh )
T-Lymphocytes, Cytotoxic ( mesh )
T-Lymphocytes, Suppressor-Effector ( mesh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida.
Bibliography:
Includes bibliographical references (leaves 235-243).
Additional Physical Form:
Also available online.
General Note:
Photocopy of typescript.
General Note:
Vita.
Statement of Responsibility:
by Martin Robert Jadus.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Martin Robert Jadus. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
10192222 ( OCLC )
029406525 ( ALEPH )

Downloads

This item has the following downloads:


Full Text














THE IMMUNOBIOLOGY OF GRAFT VERSUS HOST DISEASE AND ITS
ATTEMPTED PREVENTION USING NATURALLY OCCURRING SUPPRESSOR FACTORS








BY

MARTIN ROBERT JADUS







A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA

1983















ACKNOWLEDGEMENTS

I would like to express my appreciation to the many people who helped me complete this dissertation.

To my parents, without whose help and understanding I would never have made it this far, I extend my gratitude.

I would like to express my deep thanks to Dr. Ammon

Peck for his help and guidance through this work, especially his willingness to give me a free hand performing these experiments, related or unrelated to this dissertation which allowed me to satisfy my insatiable curiosity. In addition, his friendship is warmly remembered.

Appreciation is also given to Drs. Ward Wakeland and Art Kimura for their discussions and insights about the immunogenetics of the mouse system. Additional thanks are given to Dr. Wakeland and Vicki Henson for their preparation of monoclonal antibodies and Dr. Kimura for the column chromatography.

Thanks are also given to Dr. Paul Klein for whom a

generous supply of materials allowed me to get through the first year of this project and Drs. Richard Smith and Shiro Shimuzu from whom I was able to acquire enough interleukin 2 in order to maintain and expand the cytotoxic T cells which were important to this study. Thanks are also extended to Dr. Micheal Norcross for helping with the photography.
ii













Thanks are also extended to Drs. Roy Weiner and Noel

McClaren who shared their help and time while participating on my advisory committee.

Another round of appreciation is given to those people who helped with the mice: Mr. Lucetta for breeding the necessary mice for most of this work, Dr. Bobby Collins for his preparation of the histology of the various tissues, as well as Steve Noga for his help in interpreting the various histological and pathological sections which were reported.

Thanks are also given to Dan Cook and Dr. Jian Xiang Huang who helped me with the word processing of this work. Without their help this epic manuscript would have cost a fortune to produce.

A special round of gratitude is given to the excellent technical help of Laura Prall, Jean Mahlman and Mary Ann Searle who helped me with the arduous and time consuming tasks such as preparations of various materials which allowed me the necessary time to plan further GVH conquests. Thanks also go to Vi Sudipong for her participation with the minor histocompatibility work during the summer of 1982.

Finally, a word of gratitude is given to my fellow

graduate students: Vicki Henson, Ted Hall, Kim Peeler and Dave Shaut who had to share in my moods when things were not correct.




iii
















TABLE OF 0NTENTS

page

ACKNOWLEDGEMES ----------------------ii

LIST OF TABLES--- .------------- -----vii

LIST OF FIGURFS---------------- --x

ABSRAC -----------------------xiv

INTRODUCTION
1.1 The Graft Versus Host Reaction- ---------1
A. Sytemic -----------------------5
B. Local -------------------------6
1.2 The In Vitro Measurement of the GVH Reaction----9 1.3 Ccntrolling the Activities of T Cells In Vitro-----20 1.4 Suppressor Cells ---------------------21
1.5 The Rationale for These Experiments---- ----23

MATERIALS AN) METHODS
2.1 Animals--- ------------------ ----- 24
2.2 Antisera- ----------------------25
2.3 Canplement Dependent Antibody Cytotoxicity------ 26 2.4 Fluorescent Microscopic Determination of Cells-----26 2.5 Cell Preparations------------- ------ 27
2.6 Primary Mixed Lymphocyte Reaction------ ----- 27
2.7 Induction of Graft Versus Host Reactivity--- 28 2.8 Primed Lymphocyte Typing Tests --------------- 28
2.9 Cell Mediated Lympholysis Assays-------- ----29
2.10 Preparation of Purified Interleukin 2-----------30
2.11 Preparation of Newborn Spleen Cells -- --------31
2.12 Preparation of Alpha Fetoprotein-------------- 32


RESULTS
3.1 Inability of Allogeneic Cells to Induce GVHD in Normal Adult F1 Mice------ ----------- -----33
3.2 Generation of GJHD in Adult F1 Mice Imunosuppressed Through Irradiation- ------------ --------34
A Study of the survival rates of mice lethally or
sublethally irradiated----------- ---34
B Sublethally irradiated host provide an environment
for GVHD ------------ 36



iv












ONTENTS-Continued

C Histology of sublethally irradiated (650r) mice--40 D Pathology of GVHD in MHC disparate strains----40
E The effect of the host's age upon generation of
GVHD------------- ------ -------41
F Histopathological examination of whole H-2 disparate
G HD revealed marked effects-------- ------ 46
G Pathology of K/D or I region disparate GVHD----53
H Cellular composition of the host organs undergoing
GVHD- --------------------------60
I Recovery of viable lymphocytes from various
histoinccnpatible differences------- --- 61
J Functional activities of leukocytes obtained from
GVH animals
i mixed lymphocyte reaction/primed lymphocyte
test------- -------------------- ---- 63
ii cell mediated lympholysis assays---------- 96
K Ability of primed lymphocytes to cause mortality
in sublethally irradiated mice--------------136
L Histology and pathology of secondary disease---141
M Mortality induced by anti-I-A4 long term cultured
T cell lines and clones--------- ------150
3.3 Attempts to Modify GVIHD
A Attempts to prevent lethal GVHD using concentrated
monoclonal antibody directed towards the host's
I-A molecule ------------------- -----154
B Attempts to prevent lethal GVHD using neonatal
splenocytes
i. Inhibition of acute lethal GVHD using CBA/J
newborn spleen cells---------------157
ii. Histopathology of the experimental and control
host animals-----------------------158
iii. Functional reactivity of donor cell populations
after initial period of culturing--------161
iv. Functional reactivity of lymphocytes derived
from long term surviving host mice-------167
v. Genetic restrictions in the ability of CBA/J
newborn cells to suppress lethal GVHD----175
vi. The presence of newborn spleen cells incapable
of suppressing the GVH reactivity of adult
cells fails to modify the response of the
sensitized donor cells-------------- 178








V













CONTENTS-Continued

C Attempts to prevent GVHD using newborn spleen
supernates
i. Characterization of the newborn supernate--178
ii. Size profile of the newborn supernate
factors------------- ------------- -180
iii. Time of addition studies of the newborn
factors ----------------190
iv. Effects of the supernate in GVH------193
D Attempts to prevent lethal GVHD using AFP -----196

Discussion
4.1 The Need for Imnunosuppression for the Development of GVHID---------------201
4.2 The Kinetics of GVHD ------------------------ 202
4.3 The Functional Activities of the T lymphocytes Recovered from GVHD Animals------------------ 203
4.4 The Ncnspecific Factors Influencing Mortality in G ---D----- ------------------------216
4.5 The Attempts to Prevent GVHD Using Anti-Host I-A Antibody------ -----------------------221
4.6 The Attempts to Prevent GVHD using Newborn Suppressor Cells--------------------- ---- --------222

REFEENCES ----- ------------------------ --235

BIOGRAPHICAL SKECH-------------vi















LIST OF TABLES

Table page


1. Strength of MLR and GVHR across various regions of the
H-2.--------------------15

2. The effect of the route of injection in order to generate
primed lymphocytes in the Bl0.BR anti-(Bl0 x BIO.BR)F1
reaction. --- 35.....-------- ----1.....

3. The effect of radiation on the host in order to generate
specific primed lymphocytes in the BI0.BR anti-(B10 x
B10 .BR)F1 reaction.---- -----------------------39

4. Summary of the number of recovered cells obtained from the
GVH spleens on day 5 of the reaction.------ ------62

5. Inability of GVH primed cells to respond to mitogens.---67

6. Canparison of PLT using BlO.BR anti-(Bl0 x B1O.BR)F1
cells generated either in MER or in GH.------------ 69

7. GVH primed cells fail to alter MER primed cell responses.
------------ ----------------------- 71

8. Absorption of MLR primed lymphocytes fails to remove
proliferative responses towards K/D antigens.----------73

9. The PLT activity of GlR primed cells after treatment with
various monoclonal antibodies plus complement. --------74 10. Comparison of PLT of B6 anti-BALB/c cells generated either
in MLR or in GH. --------- --------------76

U1. Comparison of PLT of Bl0.RIII anti-BlO.A(5R) cells
generated either in MLR or in GH. --------------.78

12. Canparison of PLT of anti-Bl0.M(17R) reactive cells
generated either in MLR or in G. --------------80

13. Caomparison of PLT of anti-Bl0.GD reactive cells generated
either in MNR or in GH. --------------------------- -82

14. Comparison of PLT anti-H-2I region reactive cells generated
either in MAR or in GVH. -------85 15. Comparison of PLT of Bl0.MER anti-(A.TL x B0O.MBR)F1
cells generated either in MLR or in GH. ---------- -87
vii













TABLES-Continued

16. Canmparison of PLT of BALB/c anti-DBA/2 cells generated
either in MLR or in GVH. ----------- ------89

17. Comparison of PLT responses of CBA/Ca anti-AKR cells
generated either in MR or in GVH. ----------------91

18. PLT responses of anti-DBA/2 primed cells generated in MLR.
-----------------------------92

19. Summary of minor histocompatibility in vitro assays in
inbred H-2 mice.------------------------94

20. CML reactivity of primed BIO.BR anti-(Bl0 x B1O.BR)F1
cells generated either in MER or in GVH.------- -----97

21. CML reactivity of primed B6 anti-BALB/c cells generated
either in MER or in GVH. ------ ----- --- -99

22. CML reactivity of primed BI0.MBR anti-(A.TL x B1O.MBR)F
cells generated either in MER or in GVH. -----------101

23. CML reactivity of GH primed B10.MBR anti-(A.TL x
B1O.MBR)F supplemented in vivo with or without
interleukin 2.---------------------- -------106

24. CML reactivity of GVH primed Bl0.M(17R) anti-A/J cells
supplemented in vivo with or without interleukin 2.-----107

25. CML reactivity of BI0.AQR anti-(BlO.T(6R) x B1O.AQR)F1
cells generated either in MR or in GVH.- 109

26. CML reactivity of GJH primed B10.AQR anti-(Bl0O.T(6R) x
Bl0.AQR)F1 cells expanded in vitro with interleukin 2.---110

27. The inability of monoclonal antibodies to block killing of
B10.AQR LPS blasts by GVH primed B10.AQR anti-(B10.T(6R) x
BI0.AQR)F1 cells expanded in vitro with interleukin 2.
114

28. CML reactivity of GVH primed B10.AQR anti-(Bl0.T(6R) x
BlO.AQR)F1 cells expanded with interleukin 2.-------- 115

29. 0ML reactivity of GVH primed (B10 x B1O.Q)F1 anti-B10.MBR
cells expanded with interleukin 2. --------- ----121

30. CML reactivity of GVH primed (Bl0.MBR x B1O.GD)F1 anti-B6
cells expanded with interleukin 2. ----------- --- 125

viii












'BLES-Continued

31. Lack of OCL reactivity of GVH primed B10.HTT anti-(Bl0O.TL x
B10.HTT)F cells expanded with interleukin 2.---- --128

32. CML reactivity of primed BALB/c anti-DBA/2 cells generated
in GVH. -------------------------- ------132

33. CML reactivity of primed BALB/c anti-DBA/2 cells generated
in GVH. -------------------- --------------133

34. CML reactivity of primed CBA/Ca anti-AKR cells generated in
G -H-. ------------------------- 137

35. Suppression of lethal GVHD by newborn spleen cells in (Bl0 x
BIO.BR)F1 hosts reconstituted with seni-allogeneic adult
BIO. BR cells. -------------- -----------------159

36. Suppression of lethal GVHD by newborn spleen cells in B6 host
animals reconstituted with allogeneic adult BIO .BR cells-160

37. The ability of B1O.BR splenocytes incubated with CBA/J
newborn splenocytes to respond to mitogens.----------166

38. The inability of lymphocytes obtained fran newborn CBA/J
suppressed GVHD to respond in a PLT.-------------168

39. Responsiveness of splenocytes from CBA/J newborn mediated
acute GVqHD suppressed mice to treact to various stimulants.
-------------1----------------------69

40. The inability of cells obtained from a newborn CBA/J
suppressed GVHD to respond in a CML. ------------------171

41. The genetic restrictions in the suppression of lethal GVHD by
newborn spleen cells in host animals reconstituted with adult allogeneic spleen cells-----------------------176

42. The ability of lymphocytes obtained fran newborn SWR
splenocytes supplemanented GVHD to respond in a PLT.------- 179

43. The suppression of primary MLR by adding newborn supernate
factors at different times. -------------------- 191








ix













'IBLES-Continued

44. The effect of various substances on a primary mixed
lymphocyte reaction. ------------- ---------192

45. Lack of GVHD suppression in AFP treated (BI0 x B10.BR)F1
mice. ----------------------------198

46. Inabibility of pregnant females to be suppressed from acute
GVHD. ----------------------------- 200









































x















LIST CF FIGURES

Figure page

1. The laws of transplantation. ------------ --------------4

2. Major histoaompatibility complex of the mouse.------ --1 3. GVHR across different histocompatibility loci.---------- 14

4. Effect of irradiation upon mouse survival.------ 38 5. Lethal GVHD across major histocompatibility loci.-------43

6. The lack of a correlation of the host's age to develop
lethal GVHD in a major histocompatibility mismatch:
BIO.BR anti-BlO.WB.-------------------------------- 45

7. Mouse liver frma either a normal animal or fran a sublethally
irradiated animal on day 10. ------------------ --------48

8. Normal mouse liver, a higher magnification.---------- 48

9. The liver of a mouse undergoing acute GVHD. ----------50

10. A higher magnification of the previous liver. ---------50

11. The intestines of a normal mouse. -----------------------52

12. The intestines of a mouse undergoing acute GHD.-------- 52 13. The spleen fran a normal nouse. --------------------------55

14. The normal spleen, a higher magnification. ------------ 55

15. The spleen of an animal undergoing GVHD. ----------------57

16. A higher magnification from the previous tissue.--------57 17. Further magnification of the previous spleen. ------- --59

18. A cytocentrifuge preparation of cells obtained from a mouse
spleen undergoing acute GVHD. ------------------------59

19. The proliferative responses of GVH primed splenocytes derived
froman an acute GVHR: Bl0.BR anti-(B10 x B10.BR)F .------- 65

20. The effect of purified IL 2 given to mice undergoing K/D
GVHD: BlO.MBR anti-(A.TL x BlO.MBR)F.----------104
xi












FIGURES-Continued

21. The effect of purified IL 2 given to mice undergoing I region
GVHD: B10.AQR anti-(B10.T(6R) x B10.AQR)F1.- --- -----U-113

22. Lethal GVHD in sublethally irradiated (BI0.T(6R) x
B10.AQR)F1.---------- -------- --------------- 118

23. Lethal GVHD across KkI-Ak or I-Ak region differences.
----------------------------------- ------120

24. Lethal GVHD in sublethally irradiated C57BL/10.-------123 25. Lethal GHD in KRk-Ak or I-Ak mismatches.
---------------------------------------127

26. Effect of minor histocompatibility antigens on survival in
DBA/2 mice.----30--------------------27. Effect of minor histocompatibility antigens on survival in
AKR/J mice reconstituted with CBA/CaH cells.---- ------135

28. Ability of 106 (BL.A(4R) x B10.GD)F anti-Bl0 primed
cells to cause mortality in sublethanly irradiated mice.--139 29. The liver of a B1O.M animal undergoing GVHD from primed
lymphocytes: BI0.RIII anti-Bl0.M.-------- --------- 143

30. The lung of the BlO.M animal examined previously.-------143 31. The lung of a normal irradiated mouse 14 days after
irradiation. ----- ------------- ------ 145

32. The lung of a B10.RIII mouse reconstituted with primed
BIO.RIII anti-Bl0.M cells, day 14.-- ----- ------ 145

33. The spleen of the animal previously examined.----------147

34. The liver of a sublethally irradiated BI0.RIII mouse
reconstituted with primed Bl0.RIII anti-Bl0O.M lymphocytes.149 35. A higher magnification of the previous liver.------------149

36. Mortality in sublethally irradiated (B10.T(6R) x Bl0.AQR)F
hosts induced by anti-I-A long term cultured cell lines
and clones.-------- --- ------------------ 152

37. Inability of concentrated (ascites) anti-I-Ak antibody
(10.2.16) to prevent lethal GVHD in sublethally irradiated
BI0.A mice reconstituted with B10.A(5R) splenocytes.-----156


xii












FIGURES-Continued

38. The skin of a CBA/J newborn suppressed mouse day 72.---163 39. The liver of a CBA/J newborn GVH suppressed mouse day 72.-165 40. A higher magnification of the previous micrograph.-----165 41. A single cell suspension prepared fran a (B10 x B10.BR)F
spleen 25 days after reconstitution with BIO.BR cells an
newborn splenocytes. -------- --174

42. Dose response kinetics of various suppressor factors on
BIO.BR anti-C57BL/6 primary MLR. --------- ------ 182

43. Profile of suppressor activity fran newborn supernate passed
through a Sephacryl 300 column to inhibit a primary mixed
lymphocyte reaction: BlO.BR anti-BlO.D2. ------ ----185

44. Ouchterlony analysis of purified mouse AFP. ----------187

45. SDS-acrylamide analysis of the newborn factors.----- 189 46. Lethal GVHD in irradiated (C57BL/6 x BALB/c)F1 mice
reconstituted with BALB/c splenocytes. --------------195

47. Proposed model for GVH morbidity---------------- 234























xiii















Abstract of Dissertation Presented to the Graduate
Council of the University of Florida in Partial
Fulfillment of the Requirements for the Degree of Doctor of Philosophy


THE IMMUNOBIOLOGY OF GRAFT VERSUS HOST DISEASE AND ITS
ATTEMPTED PREVENTION USING NATURALLY OCCURRING SUPPRESSOR SUBSTANCES.


By

Martin Robert Jadus


August 1983


Chairman: Dr. Ammon B. Peck Major Department: Pathology


A model for the murine graft versus host (GVH) reaction has been developed in order to examine the potential value of naturally occurring suppressor cells and factors in preventing lethal disease. GVH reactivity was induced in sublethally irradiated adult mice by reconstituting the hosts with allogeneic splenocytes.'Nonsuppressed host/donor combinations with genetic differences at H-2 died within 14 days. Combinations with I and K/D genetic differences showed longer survival times, while mismatches involving non-H-2










xiv













generally had no mortality. T lymphocytes specifically reactive against host tissue antigens could be recovered from diseased organs such as spleen, lymph nodes and liver. Based on cell surface phenotype and in vitro reactivity, two distinct populations of cells were recovered: one proliferated against class II molecules, the other lysed cells expressing class I molecules. This situation proved different from reactivity profiles of cells activated in mixed lymphocyte reactions.

The presence of cytotoxic T cells correlated with GVHD mortality. The presence of large numbers of CTLs was common in H-2 GVH mismatches and no doubt contributed to the rapid death. In contrast, only weak, but transient CTL activity was found in mismatches involving class II antigens; however, the recovered cells could be expanded in vitro with interleukin 2. The resulting cells exhibited cytotoxic activity against both donor and host.

Attempts were made to prevent GVHD using cells from the spleens of newborn mice known to contain naturally occurring suppressor cell populations. Unexpectedly, only certain strains of newborn mice possessed the capacity to suppress lethal GVHD. The genetics of suppression by newborn spleen cells suggested two restrictions: first, the newborn spleen cells apparently must express the strongly stimulating Mls


xv












antigens, and second newborn spleen and adult donor cells must be histocompatible at a genetic region telomeric to I-A. No T cell reactivity could be detected up to 60 days in mice showing long term survival. During this time, these mice remained chimeras. After 60 days postengraftment, the cell and reactive phenotype of these mice returned to that of the host.





































xvi

















INTRODUCTION





1.1 The Graft Versus Host Reaction



The graft versus host reaction (GVH) is a unique model to study immunoregulation in that the entire sequence of an immune response (initiation, differentiation, a multifaceted effector phase) and final control is represented. An extensive body of knowledge concerning the initial events of antigenic recognition, cell types involved, cell differentiation, and final expression of immunocompetence during the GVH reaction already exists (reviewed in references 1,2). However, there is comparatively little understanding of the complex mechanisms and cellular interactions regulating the course of the on going immune reaction(s).












1







2




The most common forms of GVH reactions are runt disease, secondary disease, parabiosis and F1 hybrid disease. Runt disease occurs when mature competent allogeneic cells are injected into an immunoincompetent newborn. Growth of the newborn is inhibited and quite frequently the animal dies manifesting severe diarrhea, dermititis, hepatomegaly, and splenomegaly.

Secondary disease is seen in those adult animals who have been immunocompromised by either drugs or irradiation and have been reconstituted with allogeneic stem cells with the following results: diarrhea, dermatitis, renal lesions (immune complex deposition), hemolytic anemia and hepato and splenomegaly.

Parabiotic intoxication occurs when two allogeneic

adult mice have been surgically treated so that they share a common circulation and their lymphocytes are free to attack the other animal.

Hybrid disease happens when a Fl hybrid has been

injected with cells from the parental strain. The laws of transplantation (See Figure 1) state that when two histocompatible homozygous animals mate, the F1 progeny possess the histocompatibility antigens of both parents and are able to tolerate grafts from either parent, while the parent rejects grafts from the progeny because of the histoincompatible antigens. Thus the donor cells recognize













Figure 1. The laws of transplantation. Each inbred mouse strain is hcnozygous for the
H-2 complex. Both sets of products are expressed in the heterozygous F1 hybrids.
These mice therefore produce H-2 antigens that stimulate an immune response in
either parental strain, whereas, neither parental strain express H-2 antigens-----.
foreign to the F1 hybrid.

1) Grafts within an inbred strain (syngeneic grafts) succeed.

2) Grafts between different inbred strains (allogeneic grafts) fail.

3) Grafts fran either inbred parent strain to the F1 hybrid succeed but grafts in the
reverse direction fail.

4) Grafts fran F2 or subsequent F generations to F1 hybrids succeed.

5) Grafts from either inbred parent strain succeed in some members of an F2
generation but fail in others. Also, grafts fron one inbred strain succeed in some
members and fail in others of a backcross produced by crossing the Fl hybrid to
the opposite parent strain.

Taken from Klein (3) and Hood et al. (4)









~AA reject BB
reject parental S.H'generation
H-2a H-2b

H-2a H-2b Chromosome 17 0OO/


F,


H-2a H-2a H-2b H-2b






H-2a H.2b H-2a H-2b H-2a H-2a H-2b H-2b












the F1 being foreign, but the F1 can not normally react to the donor cells because they see those cells as being self. The advantage of this system is that the recipient possesses an intact immune system.

Currently GVH reactions are studied either systemically or locally. Both sets of reactions have been reviewed by Grebe and Streilein (2) and can be summarized as follows:



A) Systemic

Inhibition of syngeneic hematopoeitic colony forming units: Irradiated Fl mice are given parental lymphoid cells along with syngeneic bone marrow cells. The amount of erythroid cell growth is assayed by the amount of 59Fe that is incorporated into the spleen. The less 59Fe incorporated, the greater the immune response was towards the host.

The Simonsen spleen index assay: This assay utilizies the fact that when allogeneic cells are injected into a recipient the spleen enlarges in response to the allogeneic challenge.

The phagocytic index: This is an indirect test used in F1 animals 2 weeks after the parental cells are injected. Because of the increased lymphoid activity, colloidal carbon is cleared from the system in a much shorter time.








6



The focal periportal infiltration method: This system deals with enumerating the number of foci seen in the F1 liver after parental cells are injected.

The splenic explant assay: Single cell suspensions of parental cells are placed over diced F1 spleens. The culture is allowed to procede 5 days and then the culture is examined for increased physical masses.



B) Local

The epidermolytic reaction: sensitized lymph node cells are injected intracutaneously into F1 animals. The blood cells of the host are destroyed followed by nonspecific vascular destruction in the skin with noticeable epidermal necrosis.

Intraocular or intrarenal GVH: allogeneic cells are injected either into the anterior chamber of the eye or under the renal capsule; within a short period of time immune reactions cause gross morphological changes in these organs.

The popliteal lymph node assay: this assay is based on the same theory as the Simonsen spleen index except reactive allogeneic cells are injected into the animals footpad with subsequent measurements of the popliteal lymph node 1 to 2 weeks later.








7



Often, the final result of GVH is the death of the host. At times, however, the host may survive GVH, suggesting the development of "tolerance". The mechanism(s) underlying tolerance, both host towards the graft as well as graft toward the host, remain only speculative, but may result from

1] blocking factors, such as antibody possessing the potential to bind to antigenic sites of the host cells (5,6) 2] suppressor cells, which inhibit the regulating T helper cells (7,8) or

3] destruction of the stimulating components of the host, presumably the lymphoid associated cells (9,10).

The lymphoid cells of the host have been implicated in the histological and clinical manifestations of GVH in two ways. First, these cells no doubt provide the immunogenic stimuli for GVH through the trapping the donor cells within the lymphoid organs, thereby stimulating the donor cells into growth (11,12). Second, the host cells, while under GVH siege, may be nonspecifically stimulated to grow (13,14). These proliferating host cells could either be target cells or they could become autoimmune to their self via an allogeneic effect mechanism (15,16) or via an inflammatory response.

Attempts to alter GVH reactions have intensified with the development of human transplantation systems.







8




Pharmacological agents have seen a major trend towards limiting GVH reactions. Cyclophosphamide and the cortical steroids have been tried clinically with less than expected results (17,18,19). A major problem with such drugs is that these agents nonspecifically depress the immune system, thereby promoting other deleterious effects such as increased risk of infections and lymphomas. Cyclosporin A has recently become a center of a pharmacological approach to limit GVH, but it too may be nonspecifically immunosuppressive (20,21,22).

Another approach to this question has been to stimulate suppressor cells using the plant lectin, concanavalin A (Con A)(23,24,25,26). Certain doses of Con A are known to be mitogenic while other doses are known to be suppressive

(25). Unfortunately, this drug also has toxic side effects and therefore could be hard to judge adequate dosages for any given individual.

Treatment of donor cells with various antibodies and complement has been another effort which has been examined

(27). Until quite recently, the only antisera which had been employed in humans was rabbit anti-lymphocyte antisera. The hope here was to eliminate immunocompetent T cells (while leaving virgin bone marrow cells intact) before they have had a chance to become stimulated. Unfortunately, this regimen has had only limited success which could be due to








9




the poor specificity and titer of the antisera used. Perhaps the development of appropriate monoclonal antibodies might circumvent this problem (28,29). In any event, there is little disagreement on the need for better methods to achieve the final goal of tolerance.



1.2 The In Vitro Measurement of the GVH Reaction



The murine major histocompatibility complex (H-2) consists of a number of immunologically important loci: K,D,S and I (Figure 2). The K and D regions code for molecules which direct lympholysis; the S region produces serum proteins such as the fourth component of complement, while the I region genes and associated antigens have been reported to be involved in a large number of immunological phenomena such as T-B cell cooperation (30), antigen presentation by macrophage (31), helper factor (32,33), suppressor factor (34,35) and blastogenesis in mixed lymphocyte reactions, MLR (36).

Genetic disparity between donor and recipient which apparently controls initiation of GVH is determined primarily within the major histocompatibility complex (MHC) of the species, in particular the I region (in the mouse) or its equivalent in other species. Representative data



























Figure 2. The major histocampatibilty complex of the mouse. This schematic drawing of the MRC of the mouse (H-2). The H-2 is composed of several loci, as denoted on the figure. The MHC gene products is composed of several types of molecules. Class I molecules such as: K, D and L molecules are single 45,000 nolecular weight proteins noncovalently associated with beta two microglobulin molecules. The class II molecules: I-A and I-E consist of two noncovalently linked proteins of 35,000 and 28,000 daltons. Finally, the class III molecule encoded by the S region describes a serum component C'4 which is a component of the complement series of protein.






















membrane









Chromosome 17 K J E C S D L QA !1 L.


















membrane membrane Chromosomef a .








12



revealing the importance of the I region, taken from Klein and Park (37), is presented in Figure 3.

Even though K and D region differences produce some splenic enlargement as evidenced by an increased splenic index of 1.5, this increase of 1.5 probably represents just the homing in of the injected spleen cells into the host spleen without any significant reaction. But when I region differences occur, the spleen doubles or triples in size due to the development of a significant reaction.

The in vitro measurement of GVH is normally performed using the mixed lymphocyte reaction (MLR) (38,39,40,41). This test measures the proliferative phase of T cell activation following recognition of allogeneic antigens. Representative data comparing MLR and GVH are presented in Table 1 (37).

In vitro studies using the MLR as a model for GVH

have provided extensive insight into T cell recognition, differentiation, and cell interactions. Activation of T lymphocytes by MHC incompatibility results in increased DNA, RNA and protein synthesis, increased energy utilization and increased size (42,43,44). During this stage, the T cells proliferate and differentiate into either the proliferating helper T cell or the poised precursor of the cytotoxic T cell. The poised cytotoxic T cell further differentiates to































Figure 3. Graft versus host reactions across various H-2 differences. This figure represents splenomegaly indices of various mice undergoing graft versus host reactions. Taken from Klein and Park (37).








14


No. Meon Spleen INDEX and SD H-2 Region Difference of 1.0 1.5 2.0 2.5 3.0 Strain Combinoton K I Ss-S D Anim's , 810. A -BIO 17/27 AOR 5R M = = 16/32 4R B--I--* O W O ] 15/19 2R- 4R [COM~-l 17/28 BIO.A-- 4R (i] W M 20/220 BIO.A ---- ZR = j 20/30 -S-Difference in Region of H-2b Origin I *No Difference in Given Region
GVH R across 11-26 differences. Number of animals: control/experimental.
No. Mean Spleen .Index and SD H-2 Region Difference of 1.0 1.5 2.0 2.5 3.0 Strain Combination K Ir Ss-Slo D Anim's I I I I 11 I 810.R- BIO.D2 12/17 2R --- O.DZ2 I 13/19 BIO.A-+ 810.D2 = lILj 22/29 2R---- BIO.HTG[ M J ] 20/26 --0-BIO.AKM-BIOA IM l t9/20l g 2R ----* BIO.A I l l 18M/2.1

Difference in Region of H-2d Origin 'No Difference in Given Region
GVH R across 11-2 differences. Numbcr of animals: control'exlCprimcntal.

No.. Meon Spleen Index and S 0 H-2 Region Difference of 1.0 1.5 2.0 2.5 3.0 Strain Combinotion K Ir Ss-Slo D Anim'so, 1 1 I O 25 3 810.D2- 80 BR 16/25 0-AOR BI BR l ] 14/27 8002--,80 A I I51= 1/20
810 HTG-2R l --1 -18,/22
810 4R l EII II= ,23 -e-2
AOR --* I.A = I -L L] 17/29
BIO.A- 810 BR -L l 20/21
BeOAKM-BI BR [I J L--"20/2-7 t

Difference in Region of H-2k Origin i = No Difference in Given Region
(CVII R across 11-2L differences. Number of animals: control/experimcntal.

No. Mean Spleen Index and SD H-2 Region Difference of 1.0 1.5 2.0 2.5 3.0 Slroin Combinoaton K Ir Ss-Sto 0 Anim'si 1 1 a a a I, 1 ,I
BO 81---ZBG 15/21
e10.A-- GR 1 18/2I 5
SOR-- eO1.G I I 20/27-O
10.A- *AOR l-- Li I- 19/25 SR -* BIO.G I 1 l -1 14/23


Difference in Region of H-2QOrigin No Difference in Given Region
GVII R across 11-.2 differences. Number of animals: control/experi.tental.








15







Table 1.
Strenth of MLR and GVHR across various regions of the H-2.

MLR GVHR
H-2 region average ratio range mean spleen range difference of stimulation index K 2.0 .0.7-4.7 1.4 1.2-1.6
D 1.8 0.8-5.4 1.4 1.2-1.7 I-A 6.4 3.7-9.2 1.8 1.8 I-B+S 2.7 1.1-4.4 1.5 1.1-1.8 I+S 5.8 2.7-12.8 2.6 2.6 K+I 6.6 3.2-18.3 2.8 2.6-3.1 D+S 2.0 0.7-4.7 1.5 1.3-1.6 K+D 3.4 3.0-3.8 1.8 1.1-2.5 K+S+D 3.3 1.5-8.6 2.3 2.2-2.4 K+I+S+D 7.2 1.2-33.6 2.8 2.4-3.1


Taken from Klein (3)








16



the mature killer cell after receiving the help signal from the helper T cell.

Killer T cells were originally characterized as possessing the Lyt 2 and Lyt 3 antigens by negatively selecting T cells with anti-Lyt 1 antisera and complement (45,46). Recently, killer cells possessing the Lyt 1 phenotype have been described (47). The discrepancy in 2 killer T cell phenotypes may be due to the sensitivity in the antisera used. Originally anti-Lyt 1 antibody and complement killed the high density Lyt 1+ cells so that weakly positive Lyt 1+ cells still are present. Only through sensitive immunoflourescent techniques can one observe the low density Lyt 1+ cells (48).

Helper T cells reacting against whole haplotype, I

region, or non-MHC antigens have been shown to bear only Lyt 1 antigens (49,50). In contrast, T cells responding against K/D differences have been found to be Lyt 1,2,3+ cells (51). This as well may be an artifact with the Lyt 1,2,3+ cells being precursors to the T helper cells, Lyt 1+.

The blast cells do not remain in this activated mode for long as they revert to small lymphocytes capable of exhibiting memory (52). These cells can be restimulated by the initial antigen in a secondary manner, with the peak response occurring on day 2 of culture as opposed to days 5 and 6 found in a primary reaction. The primed lymphocyte








17



test (PLT) utilizies this concept of in vitro restimulation to specifically quantitate similar histocompatible antigens on third party cells (53,54). This procedure has only been developed in the last few years; however, the principles of these in vitro restimulations offer much potential in probing some basic questions of vital importance in immunobiology such as the activation of memory cells, the differences between primary and secondary stimulation, the specificity of the response towards antigens, and the number of different cell subpopulations responding towards one antigen. All these activities have been observed at the gross level, but the actions of single clonally expanded cells have only been recently approached. By selectively expanding blast cell cultures one may be able to examine the fine specific activities of individual cells which were not available previously.

T cell clones have been established in various

laboratories using a variety of different strains of mice as the source of responder cells. Essentially two types of functional clones have been found. One type of clone is cytotoxic towards the stimulating cell only when that specific antigen bearing cell and some exogenous helper factor(s) are present (55,56,57,58,59). Since these clones carry out this cytotoxic effector function, it is presumed that these cells originated from a cytotoxic T cell








18



possessing Lyt 2,3 antigens. These cells may be maintained in culture for weeks solely by the presence of the exogenous factors even without the initial stimulating antigen. Interleukin 2 (IL 2) is one of those exogenous factors; it is a 30,000 to 35,000 dalton protein derived from splenic lymphocyte cultures stimulated by T cell mitogens Con A or phytohemagglutinin (60,61,62,63,64), from other MLR reacting supernates (65,66,67) or from clones of T helper cells (58,59,68).

The second type of clone undergoes proliferation when presented with the specific priming antigen, e.g., histoincompatible cells, or soluble antigens, but does not exhibit any cytotoxic potential. No exogenous factors are apparently needed by these clones. However, these cells can become addicted to IL 2 and can lose their specificity and ability to respond to antigen. A few of these clones have been shown to produce factors which activate the cytotoxic T cell clones in vitro (58,59), strongly suggesting that these clones are members of the helper T cell class. Helper T cell clones have also been shown to function in vivo by increasing antibody formation towards T cell dependent antigens, e.g., sheep red blood cells and horse red blood cells in nude mice (69).

In addition to nonspecific helper factors such as IL 2, T cells have been claimed to produce antigen specific








19



factors which stimulate naive B cells into making specific antibodies towards a given antigen. Some of these factors bind to the antigen directly, possess Ia determinants, and have a molecular weight of 35,000 daltons. These factors when isolated from T cells can direct the B cells into antibody secretion (70,71,72,73). Whether a given T cell secretes both IL 2 and antigen specific helper factor is not known.

The majority of cytotoxic clones which have been established have activity against specific membrane moieties, like H-2K or H-Y antigens (74,75,76). In contrast, the majority of proliferative helper T cell clones have activity against non-MHC antigens (77,78), although a few have been selected with activity against H-2 I region gene products (Ia) (79,80).

One of the problems with long term lymphocyte cultures and clones has been the frequent and regular appearance of a crisis phase in the growth pattern of these cells (68). After 3 to 6 months in culture the majority of the responding cells die; the remaining cells replicate at a slower pace than they did before crisis set in. This crisis period lasts for about 3 weeks, after which time the cells may once again grow. Thus, long term cultured cells can be used for analysis only during the limited periods between crises.








20



1.3 Controlling the Activities of T Cells In Vitro



Alpha fetoprotein (AFP), a normal component of fetal and newborn sera, has been shown in both human and murine systems to exert selective suppressive effects on various functions of T lymphocytes, including T cell dependent antibody synthesis, T cell mitogenic responsiveness, and T cell mediated allogeneic reactivity (81,82,83,84,85,86). In addition, recent reports have revealed that under certain circumstances AFP may exert a supportive influence on in vitro cell growth with one manifestation being the in vitro induction of suppressor T cells (83).

Analysis of the impact exerted by AFP on the

recognition and subsequent proliferation of T lymphocytes reacting in MLR against defined histocompatibility alloantigens has revealed a highly selective activity in the suppression of lymphocyte responses (81,82). In general, AFP inhibits Lyt 1+ T blast cells reacting against I region structures, including reactions against Mls locus products, but fails to inhibit Lyt 2+ cells stimulated by K/D alloantigens (81). Thus, it seems clear that AFP exerts its suppressive activity in MLR via selective interference with I region triggering systems. However, AFP also suppresses the effector phase of the T cell mediated cytotoxic reaction, thus suggesting a broader spectrum of regulatory








21



activity. For example, AFP, when present during the primary activation phase of T cell responses, not only suppresses the subsequent in vitro generation of effective cytotoxic T cells in strain combinations with I plus K/D region differences, but also in strain combinations possessing only K/D region differences where the proliferative phase was unaffected. If AFP interferes only with I region triggering, then it would have been expected that at least in reactions directed against isolated MHC SD region associated gene products not only the proliferative but also the cytotoxic phase would have remained refractive to the suppressive activity of AFP. More recently, studies by Peck et al.

(87) have shown that AFP acts on the stimulating cell population known to initiate T cell reactivity. Furthermore, the T cell subpopulation which is refractive to AFP could be shown to exert suppression of normal primary responses. This fact suggests that AFP may be a physiological substance which could be used to control specifically the immune response in GVH reactions.



1.4 Suppressor Cells



A subpopulation of cells has been described which

resides in the spleens of the newborn mouse which possesses a short lived antigen nonspecific suppressor activity for








22



immune reactions (88,89,90). This suppressor activity is not associated with spleen T cell populations as it is absent from purified spleen T lymphocytes, resistant to treatment with anti-I-J and anti-Ia antisera, present in the spleens of T cell deficient nude mice. In addition, suppressor activity is not due to macrophages since the effector cells fail to adhere to either plastic or Helix pomatia lectin coated plates. Natural killer cells can also be excluded since the suppressor activity fails to pass through Ig antiIg coated columns (Peck, unpublished results). In addition, the cells from those animals are capable of producing a soluble factor which is capable of inhibiting adult cell responses both MLR and CML.

'Another set of suppressor cells has been postulated (2) to be responsible for the inability to transfer GVH from a host animal undergoing GVH reaction to a second normal animal even when both animals are genetically identical. Perhaps that time required for secondary reactivity to establish itself in the second host may permit an antigen specific suppressor cell to develop.

Both of these activities need to be further

investigated as physiological entities to suppress immune responsiveness.








23



1.5 The Rationale for These Experiments



GVH reactions have been studied in mice in a wide

variety of ways, ranging from local footpad swelling assays to splenomegaly studies in newborn hosts. Unfortunately, the majority of these studies are not homologous to the situation seen in human GVH following bone marrow transplantation. Studies dealing with the homologous situations in adult mice undergoing GVH can be classified into two main categories: first, those which have dealt with the histopathological lesions, and second, those studying the ability of cytotoxic T cells to develop using an entire H-2 mismatch. The purpose of this research was to develop a model to study GVH reactions resulting from specific H-2 and non-H-2 incompatibilities between donor and host in order to study the genetic control of T lymphocyte reactivity in GVH disease, then determine the feasibility to control this reactivity with physiological pregnancy associated substances.















MATERIALS AND METHODS





2.1 Animals



Inbred lines of mice used in these studies and

maintained in the Department of Pathology, University of Florida, include A/J, AKR/J, A.TFR5, A.TL, BALB/cJ, BALB/c dm2, B1O.A, B10.AQR, B1O.A(2R), BlO.A(3R), BlO.A(4R), BlO.A(5R), B1O.BR, B1O.BUAl6, B10.CHA2, B10.D2, B10.F, B10.HTT, Bl0.M, BlO.M(17R), B10.MBR, Bl0.PL, BlO.RIII, B1O.S(7R), B1O.S(9R), B1O.SM, B1O.T(6R), B10.Q, B10.TL, CBA/CaH, CBA/J, C3H/HeJ, C57BL/6J (B6), C57BL/6 bml, C57BL/10 (B10), C57BR, C58, Dl.C, DBA/lJ, DBA/2J, PL/J, NZB, RF/J, SEA/J, SEC/J, SJL/J, SM/J and SWR/J. Breeding pairs of Bl0.GD, B1O.RIII, BlO.S(7R) and BlO.S(9R) were originally provided by Dr. Duncan, Department of Cell Biology, University of Texas, Dallas, Texas, while B10.AQR and BI0.TL were obtained from Dr. Shreffler, Department of Genetics, Washington University, St. Louis,









24








25



Missouri. B10 males were bred to B10.BR females, A.TL males bred to B10.MBR females, and Bl0.T(6R) males bred to B10.AQR females provided the Fl hybrids, (B10 x B1O.BR)F1, (A.TL x Bl0.MBR)F1, and (B1O.T(6R) x B1O.AQR)Fl. Both male and female mice ranging in age from 6 to 24 weeks were used; however, mice were sex matched when used for various experiments.



2.2 Antisera



Monoclonal anti-Lyt, anti-Thy-i and anti-I-Ak

antibodies were obtained from cell lines 53-7.313 (anti-Lyt 1), 53-6.72 (anti-Lyt 2), HO-13-4 (anti-Thy-1.2) and 10-3.6.2 (anti-I-Ak) generously provided by Dr. Ledbettor and Dr. Herzenberg via the Salk Institute Cell Distribution Center, San Diego, California, while anti-Kd (B 312) and anti-I-Ek (14-4-4) were provided by Dr. D. Sachs, NIH. Each cell line was grown at high concentration in RPMI 1640 supplemented with fetal bovine serum to 10%. Supernates were used undiluted.

The antibodies 10.2.16, B 312 and 14-4-4 used in this research were generated in (B6 x BALB/c)F1 mice using the ascites approach. These antibodies were demonstrated to high cytotoxic titers and were generously provided by Dr. E. Wakeland.








26



Arsenilic acid conjugated anti-Lyt 1 and anti-Lyt 2 antibodies and fluorescenated rabbit anti-arsenilate antibody were obtained from Becton-Dickinson, Oxnard, California. Fluorescenated rabbit anti-mouse Ig antibody was obtained from Cappel Laboratories, Cochransville, Pennsylvania.



2.3 Complement Dependent Antibody Cytotoxicity



A two step complement dependent cytotoxicity assay was used to treat cell populations with various antisera. Spleen cells at 5.0 x 106 cells/ml were incubated with antiserum for 45 min. at room temperature. The cells were then washed, resuspended in rabbit complement (Accurate Scientific, Hicksville, New York) and reincubated at 37 C for 45 min. Cell viability was assessed by trypan blue dye exclusion.



2.4 Fluorescent Microscopic Determination of Cells



Immunofluorescent staining of lymphocytes was performed by reacting 5.0 x 106 lymphocytes with either arsenilate conjugated anti-Lyt 1 or anti-Lyt 2 antibodies for 45 min. at 4 C. The cells were washed 3 times with phosphate buffered saline (PBS) followed by an incubation with fluoresceinated rabbit anti-arsenilate antibody for 45 min







27




at 4 C. The cells were examined for fluorescent staining through a phase contrast microscope equipped with a Zeiss Ploem UV illuminator. Cell surface immunoglobulin was detected in a similiar manner using a one step incubation with fluorescein conjugated rabbit antimouse immunoglobulin.



2.5 Cell Preparations



Whole spleen leukocyte populations were prepared as

described by Peck and Bach (91). In brief, spleens freshly removed from mice were dispersed by pressing the spleen through a wire mesh screen into PBS. Following one wash, the red blood cells were lysed in a 10 minute 0.84% ammonium chloride treatment. The resulting leukocytes were washed once and resuspended in PBS to appropriate cell concentrations.



2.6 Primary Mixed Leukocyte Reaction



Primary MLRs were carried out according to the protocol of Peck and Bach (91). Throughout the study, 60 x 106 splenic leukocytes were cultured together with 100 x 106 (2000R) stimulating whole spleen cells in 30 ml EHAA media supplemented with normal mouse serum to 0.5% (92). Cell cultures were harvested between days 7 to 10 of incubation








28




and examined for reactivity in secondary MLR (PLT) and cell mediated lympholysis assays (CML).



2.7 Induction of Graft Versus Host Reactivity



Forty million donor (responding) splenic leukocytes were injected intravenously via the tail vein into sublethally irradiated (650R) recipient (host) adult mice. Animals were fed on lab chow and given acidified water to drink. No major problems developed from bacterial infections.

At various time points (indicated in the text and

footnotes) the recipient mice were sacrificed. Their livers, spleens, kidneys, intestines, lungs and skin were removed, fixed in formalin and embedded in paraffin for routine hemaoxylin and eosin stained histopathological study. Cells present in the spleens, lymph nodes and livers were prepared and examined for functional activities.



2.8 Primed Lymphocyte Typing Tests



Antigen activated cells were separated from small and dead cells on Ficoll-Paque (Pharmacia Fine Chemicals, Piscataway, New Jersey) (d=1.077) density gradients. PLT assays were performed in round bottomed 96 well Lindbro








29



microtiter plates (Flow Laboratories, McLean, Virginia) as described by Peck and Wigzell (93). Cultures consisted of either 0.04 x 106 responding spleen, lymph node or liver cells cultured with 0.5 x 106 irradiated stimulating spleen cells.

At appropriate times of the secondary culture, as indicated in the Tables, 1.0 pCi tritiated thymidine (3H-TdR) (Amersham, Boston, Massachusetts) in a volume of

0.02 ml was added to each well for 12 hrs. Cells were then aspirated through Whatman glass fiber filters with a multiple sample harvestor (Otto Hiller Inc., Madison, Wisconsin) and total 3H-TdR incorporation was determined by liquid scintillation procedures. Data are expressed in counts per minute of the average of either duplicate or triplicate cultures. Standard deviations of the average are included.


2.9 Cell Mediated Lympholysis Assays



CML were performed according to the procedure detailed elsewhere (94). The effector cells were generated in primary MLR or GVH. Alloantigen activated cells were collected from mixed lymphocyte reactions or diseased organs, centrifuged and washed twice in medium containing newborn calf serum (Biocell Laboratories, Carson, California).







30




Concanavalin A (Pharmacia Fine Chemicals) stimulated

spleen cells were used as the target cells. Approximately 36 hrs before the CML assay, appropriate target cell cultures containing 12 x 106 spleen cells in 5.0 ml of EHAA with 10% newborn calf serum were established in culture dishes. Target cells were incubated 2 hrs with 400-500 pCi Na2 51Cr04. The labeled cells were washed 3 times in fresh medium.

Cell destruction was performed in V bottomed microtiter plates (Lindbro) using various effector cell numbers plus

1.0 x 104 labeled target cells. Cell destruction proceeded 6 hrs at 37 C after which time the plates were centrifuged, the supernate collected and the quantity of released of 51Cr determined. Percent cytotoxicity is expressed as 51 51 51Cr released (experiment)-5-Cr released (spont.) xl00
51Cr released (maximum) -51Cr released (maximum)

Spontaneous release ranged between 15 to 20% of the maximum release.



2.10 Preparation of Purified Interleukin 2



Interleukin 2 was produced by culturing EL-4 G-12

cells, an azogaunine resistant EL-4 (T lymphoma cell line) with 8 pg/ml of concanavalin A for 18 hrs as described by Shimizu et al. (95). The interleukin 2 was collected and








31



precipitated by ammonium sulfate and subsequently dialyzed in PBS. The supernate was concentrated and passed through a G-200 column. The fractions containing the IL 2 were collected and concentrated. These samples were then sterile filtered. The activity of the preparation was tested by using an IL 2 dependent cell line by Dr. Shimizu.



2.11 Preparation of Newborn Spleen Cells



Newborn mice 1 to 3 days old were sacrificed; their

spleens were removed using sterile techniques. The spleens were passed through a fine wire mesh screen to get a single cell suspension. The cells were washed once with PBS and then cultured in EHAA media at a concentration of 10 x 106 cells/ml. Normally 2 x 106 viable cells are obtained per spleen.

To generate suppressor cells used for the prevention of GVH disease, the newborn cells were cultured 1 day prior to the addition of the donor cells. This mixture of cells was allowed to incubate an additional day. These cells were harvested, counted and injected into sublethally irradiated host mice.








32



2.12 Preparation of Alpha Fetoprotein



Pregnant mice between (10 to 14 days) were cervically dislocated and surgically opened. The amniotic sac was punctured using a needle; the amniotic fluid was then aspirated into a collection flask. The amniotic fluid was passed through an affinity column of rabbit anti-mouse AFP antibody generously provided and established by Dr. A. Kimura. The AFP was eluted off the column using glycine-HC1 buffer pH 3.5. The AFP was subsequently dialyzed in PBS three times in 1000 x volume. The AFP was then tested in a primary MLR to insure that the collected substance had suppressor activity in it.

















RESULTS





3.1 Inability of Allogeneic Cells to Induce GVHD in Normal Adult F1 Mice


Adult male and female (B10 x B1O.BR)F1 mice were

injected either intravenously or intraperitoneally with forty million Bl0.BR splenocytes. Each set of mice was observed for a period of 1 month, during which time no signs of ill health were apparent. This experiment was repeated on a new set of mice, except this time the experiment was terminated on day 5 in order to examine the spleens of these animals. The mice which were injected intraperitoneally showed no signs of illness; the spleens of these animals looked normal and possessed a normal number of leukocytes (60 x 106 cells/spleen). In contrast, the mice which were injected intravenously exhibited splenomegaly and had twice the number of leukocytes, 120 x 106 cells/spleen. Despite this enlarged spleen, the animals looked and acted like normal mice.








33








34



The spleens of these two sets of animals were prepared into single cell suspensions and forty thousand cells were dispensed into each well of a microtiter tray. The cells were then tested with a panel consisting of irradiated spleen cells from five different mouse strains as is done in a typical primed lymphocyte typing test. The cells from mice which were injected i.p. did not significantly respond towards any of the stimulating cells (Table 2, column 1). In contrast, those cells obtained from the i.v. injected mice looked larger and appeared activated. These cells did respond towards the panel of stimulating cells (Table 2, column 2); however, it was without any specific pattern.



3.2 Generation of GVHD in Adult F1 Mice Immunosuppressed Through Irradiation A. Study of the survival rates of mice lethally or sublethally irradiated


To obtain conditions in host animals permitting

development of GVHD, it was necessary to immunosuppress the host through irradiation. A dose of irradiation was desired so that the animal would not die of infection due to the associated leukopenia, yet be compromised enough so that the donor cells would be opposed with the least possible resistance.







35









Table 2.
The effect of the route of injection in order to generate primed
lymphocytes in the BlO.BR anti-(Bl0 x BlO.BR)F1 reaction.



3H-TdR Incorporation CPM+SD (a) H-2 Genetics
Stimulator Mice injected by: (b) Strain K A J E D intraperitoneal intravenous


none 173+ 66 8570+ 107 BIO.BR k k k k k 622+298 7551+1402 B10 b b b b b 525+ 4 10875+ 341 (Bl0xBl0.BR) b b b b b
F k k k k k 679+232 8046+ 6797 BIO.T(6R) q q q q d 711+711 8651+ 75 DBA/2 d d d d d 813+ 95 9965+ 835



a) PLTs wese harvested at 48 hrs following a 12 hr pulse with 1
pCi of H-TdR.
b) Primed cells were obtained on day 5 from sublethally
irradiated (B10 x BlO.BR)F1 mice which were reconstituted
with BlO.BR splenocytes by either an intraperitoneal or
intravenous route.







36




Figure 4 shows the survival pattern of groups of 4 mice which were irradiated with either 475, 650, 875 or 1100 rads. A dose of 475 or 650 rads produced no observable detrimental effect up to 75 days post irradiation when the experiment was terminated. Higher doses of radiation used (875 or 1100) caused death in these animals by day 14. Thus, lethal irradiation induces quick mortality while sublethal irradiation has no obvious detrimental effect on the mice.



B. Sublethally irradiated hosts provide an environment for GVHD


To test the effect of irradiation on the host in order to generate primed lymphocytes, (B10 x BlO.BR)F1 host animals were divided into two groups with one group receiving 650R while the other group did not receive any irradiation. The host animals were injected i.v. with 40 x 106 B10.BR splenocytes. Five days later these animals were sacrificed, their spleens were removed, and the cells tested for alloreactivity in MLR. The results of one experiment are reported in Table 3. Cells from animals which were not irradiated produced cells which did not respond specifically towards any particular strain of mouse. However, cells from those mice which were sublethally irradiated showed patterns of reactivity suggesting specific reactivity against the





















Figure 4. Effect of irradiation upon mouse survival. This figure represents the survival curves of groups of mice composed of 4 individuals, in which each group received a different dose of irradiation: 475, 650, 875 and 1100 rads. Mortality was scored an the day the mouse died.













475 rods
100 / 650 rods


75
-J 0 875 rods 0 1100 rods S50


z
w
0 25
w


2 4 6 8 10 12 14 75 DAYS AFTER IRRADIATION










00








39








Table 3.
The effect of radiation on the host in order to generate specific
primed lymphocytes in the B10.BR anti-(B10 x BI0.BR)F1 reaction.



3H-TdR Incorporation CPM+SD (a) H-2 Genetics
Stimulator Condition of Host: (b) Strain K A J E D Not Irradiated Irradiated


none 8570+ 107 1947+ 133
BI0.BR k k k k k 7551+1402 1429+ 252 B10 b b b b b 10875+ 341 42062+1809 (B10xBl0.BR) b b b b b
F k k k k k 8046+ 697 44871+ 402 BlO.T(6R) q q q q d 8651+ 725 6146+ 292 DBA/2 d d d d d 9965+ 835 10168+2227



a) PLTs were harvested at 48 hrs following a 12 hr pulse with 1
pCi of H-TdR.
b) Primed cells were obtained on day 5 from either sublethally
irradiated or non irradiated (B10 x B10.BR)F hosts which
were reconstituted with BlO.BR splenocytes injected i.v.








40



antigens which were found on the host. Thus, by immunocompromising the host it is possible to generate a GVHR.



C. Histology of sublethally irradiated (650 rads) mice



Various tissues from animals which had been previously irradiated 10 to 15 days earlier were removed, prepared for thin sectioning and then examined for abnormalities. Liver, kidney and small intestines appeared normal. Lymphoid tissue such as thymus and spleen were atrophic. The spleen contained 0.05 x 106 cells and the histology of the spleen revealed necrosis, but did show signs of regeneration with the presence of megakaryocytes.



D. General pathology of GVHD in MHC disparate strains



Sublethally irradiated animals reconstituted with 40 x 106 allogeneic or semi-allogeneic splenocytes developed acute GVHD. As early as five days postgrafting, mice started to deteriorate physiologically: The mice became lethargic, developed hunched postures, presented a wasting appearance and developed diarrhea. Often immediately before death the mice felt hypothermic and were shivering. In semi-allogeneic combinations such as (B10 x B1O.BR)F1 mice reconstituted








41



with BlO.BR cells, mice began to die by day 7 and by day 14 to 15 all the mice had expired. In allogeneic combinations such as the B6 anti-BALB/c reaction an acclerated course of the disease was seen, and by days 5 to 6 all the mice had died.

A comparison of the survival rates due to different

histocompatibility loci is presented on Figure 5. An entire H-2 disparate GVH reaction [B6 anti-BALB/c] resulted in a very short lifespan. An I region mismatch [B10.AQR anti(BlO.T(6R) x Bl0.AQR)F1] produced a slightly delayed mortality, while a K/D disparate GVHD [B10.MBR anti-(A.TL x Bl0.MBR)F1] resulted in about half the mice surviving the first 20 days.



E. The effect of host's age upon generation of GVHD



Another study was undertaken to determine whether age of the host had any influence on the rate of GVH mortality. In one experiment an entire H-2 combination was used: B10.BR anti-Bl0O.WB. The age of the hosts ranged from 22 days (mice are normally weaned on day 21) to 70 days old (young adult). As can be seen in Figure 6, the majority of the mice died between days 4 to 8 and by day 10 all the mice had succumbed to GVHD. Thus, lethal GVHD developed similarly in mice






















Figure 5. Lethal GVHD across najor histocmnpatibility loci. Host BALB/c, (BlO.T(6R) x B10.AQR)F1 and (A.TL x Bl0.AQR)1 mice were sublethally irradiated with 650 rads, then reconstituted with 40 x 10 lallogeneic B6, Bl0.AQR, or BlO.MBR splenocytes, respectively, via the tail vein. Mortality was scored on the day the mouse died.







I00


A 4 M ice w ere sub letholly irradiated and reconstituted with 40x 106 splenocytes A Whole H-2 C57BL/6 U Bolb/c 75
Iregion BIO.AQRE(BIO. T(6R)XBIO.AQR) F,
0 K/Dregion BIO.MBRi (A.TL X BIO.MBR)F,


-.


r 5


z

cr











10 20 30 40 50 DAYS POSTENGRAFTMENT



L.























Figure 6. The lack of a correlation of the host's age to develop lethal GVHD in a major histocompatibility mismatch: BI0.BR anti-BlO.WB. B0O.WB mice of varying ages 22 days old to 70 days old were sublethally irradiated and reconstituted with 40 x 10 allogeneic BIO.BR splenocytes via the tail vein. Mortality was scored on the day the mouse died.













100

AGE OF HOST MICE A 22 DAYS OLD n 2 a 29 DAYS OLD n 6
38 DAYS OLD n -6
51 DAYS OLD n.6 75- o 70 DAYS OLD n *6






50

z
w

a.
25






I I I I I o, I I I I I 2 3 4 5 6 7 8 9 10 II 12 DAYS POSTENGRAFTMENT











Ln








46



ranging in age from 22 to 70 days old. Experiments performed in this study thus used mice which were this age.



F. Histopathological examination of whole H-2 disparate GVHD revealed marked effects


The pathology described here is similar to that

previously reported by Rappaport et al. (96). The livers of these GVH affected animals often changed from a normal red color to a pale white coloration. This condition seemed to frequently occur before any wasting syndrome presented itself. These livers showed signs of perivascular cuffing, dilation of the veins, massive necrosis with no signs of regeneration. There was marked evidence of leukocytic infiltration in the parenchyma as well as along the central veins (Figures 7 to 10). The yield of leukocytes from the GVH liver varied in different experiments but usually 20 x 103 to 70 x 104 cells/liver could be recovered. In contrast, normal livers failed to yield any significant amount of leukocytes.

Similarly, examination of the small intestine revealed drastic changes: the villi were dilated with columnar metaplasia. Exfoliation of the villi was also noted to be higher than normal. Leukocytes were devoid in the villi (Figures 11 and 12). Destruction and necrosis was obvious, and leukocytic and plasma cell invasion of the basement





























Figure 7. Mouse liver fran either a normal animal or from a sublethally irradiated animal on day 10. Both livers exhibit no drastic changes morphologically. The liver is uniformly packed with hepatocytes. Central veins are seen at lower left and center. A bile duct is seen in the upper right.

Figure 8. Normal mouse liver, a higher magnification. The previous section was examined under higher magnification. The artery is situated in the center. Notice the uniformity of the cells. Staining the cells with hematoxylin and eosin reveals the hepatocytes have an eosinophilic cytoplasm with well defined cell membranes.












48








1.' 0,







IN








9 1 A 44




*4 ji





























Figure 9. The liver of a mouse undergoing acute GVHD. This liver was obtained from a (BI0 x BI0.3R) mouse undergoing acute GVHD induced by 40 x 10 lB10.BR splenocytes at day 10 of the reaction. Perivascular cuffing exists along the central vein. Leukocytes can be seen along the vein as well as infiltrating the parenchyma. The normal architecture of the liver appears to be disturbed with numerous vacoules found in the parenchyma.

Figure 10. A higher magnification the previous liver. Leukocytes have infiltrated along the bile duct and have invaded the parenchyma. Notice the loss of normal cellular distribution. The hepatocytes show coagulative necrosis, the cytoplasm of the cells has been disrupted with only nuclear remnants left.










50







-w







Or





























lgo
i r ."~Cd





























~"C44&




























Figure 11 The intestines of a normal mouse. The intestines of a normal mouse are similar to those found in a sublethally irradiated mouse at day 10, although there is a slight decrease of (10 to 20%) leukocytes in the villi. The intestine here shows the villi are intact and have leukocytes along the lacteals.

Figure 12. The intestines of a mouse undergoing acute GVHD. A sublethally irradiated (B10 x B10.BR)F1 mouse was injected with 40 x 10 BlO.BR splenocytes and was examined on day 10 of the reaction. Epithelial cells are still along the periphery of each villus. Necrosis is observed in the epithelium. The lacteals are remarkedly devoid of leukocytes (90 to 95%).









52





























16&
FP q~












4









40



z I L.


ir i







53




membranes was pronounced. Several attempts were made to extract leukocytes from these tissues, unfortunately, no cells were recovered.

Histopathological examination of the whole H-2 GVH spleens on days 6, 10, and 15 days postgrafting revealed marked atrophy with disruption of the normal white pulp architecture, similiar to results found in the sublethally irradiated mice (Figures 13 to 17). However, more viable cells (5 to 15 x 106 cells/spleen) could be recovered from these types of spleens.



G. Pathology of I or K/D region disparate GVHD



In I region GVHD the spleen and liver appeared to have leukocytic infiltrates which progressed with time. The predominate cell type found in the spleen was the polymorphonuclear leukocyte (PMN)(70 to 90%). The intestines developed abnormalities in the second week after engraftment. The lesions which were seen in these animals were never as severe as those seen in whole H-2 disparate GVHDs. In addition, the pathology of one animal in a given series of GVHD was at times dissimilar to that seen in another animal indicating advanced stages of disease occurred in some animals but not in others. Mice died between 8 to 17 days after engraftment (Figure 5).































Figure 13. The spleen from a normal mouse. The germinal centers are slightly hypercellular, but the normal architecture is intact.

Figure 14. The normal spleen, a higher magnification. This section here displays two germinal centers.









55













Jl


ij~dt,% ~ ~ li "r.14 '






























Figure 15. The spleen of an animal undergoing acute GVHD. A sublethally irradiated (B19 x B1O.BR)F1 mouse was reconstituted with 40 x 10 B10 splenocytes. The animal was sacrificed on day 10 of the reaction. The overall architecture of the spleen has been disrupted and appears to be identical to that produced by a sublethal dose of irradiation. The red pulp appears to be repopulated by cells.

Figure 16. A higher magnification from the previous tissue. Areas are fibrosed with collagen deposition and have numerous cells in the area. A great majority of the cells appear dead and this is confirmed when a single cell suspension is examined using trypan blue dye.












rI
















P
4








'KA~






#J~c





























Figure 17. Further magnification of the previous spleen. The cellular infiltrate of this area appears to be mnononuclear in origin. Massive amounts of necrotic cells appear to be observed in the center of the field.

Figure 18. A cytocentrifuge preparation of cells obtained from a mouse spleen undergoing acute GVHD. When the GVH spleen is passed through a wire screen and the cells are passed over ficooll gradients, to remove the dead cells, the remaining viable cells appear to show a variety of different cell types. this figure shows that about 50% of the recovered cells are PMNs. Several lymphoblasts are found
along with numerous small lymphocytes.








59








60



In K/D region disparate GVHD, pathological conditions similar to those found in I region GVHD were seen; only the symptoms of the disease seemed to be delayed or absent. Death usually occurred from days 10 to 20 and some mice survived longer than four months (Figure 5). Infiltrates were found in the spleen and liver; again the predominant cell type recovered was the PMN (70 to 95%). Intestines were frequently normal; very few villi were dilated and depleted of leukocytes. However, about 60% of the animals did die in an emaciated condition.


H. Cellular composition of the host organs undergoing GVHD


Cytocentrifuge preparations of single cell suspensions of anti-H-2 GVH spleens revealed 40 to 50% of the recovered viable cells were PMNs, while 50 to 60% were lymphocyte/monocytes (Figure 18). Leukocyte preparations extracted from the livers revealed a similiar composition. Similar preparations of anti-I or anti-K/D region GVH spleens contained 70 to 90% PMNs. Culturing the leukocyte preparations overnight in tissue culture medium resulted in the majority of the PMNs dying, thereby facilitating isolation of the lymphocyte/monocyte population on ficoll-isopaque density gradients (d=1.077). Of the remaining viable cells, 30 to 35% were immunoglobulin








61



positive (determined by cell surface immunofluorescence); 55% stained for the Lyt 1 marker, while 20% stained for the Lyt 2 marker. Monoclonal anti-Thy 1.2 antibody plus complement killed 60% of the cells providing evidence that the majority of the cells were T cells bearing the Lyt 1 marker.


I. Recovery of viable lymphocytes from various histoincompatibility differences


The number of viable lymphocytes out of the spleen varied depending upon the genetic disparity. Table 4 summarizes these findings. The spleens of the hosts undergoing GVHD were removed on day 5 of the GVH reaction and were incubated overnight; the remaining viable lymphocytes were then enumerated the next day. In general, those combinations with the maximum genetic mismatch (entire H-2 mismatched reactions: combinations 1 to 3) yielded the most lymphocytes/spleen: 2.5 to 5.1 x 106. Whereas in those combinations which differed in class I molecules produced between 2.2 to 4.6 x 105 lymphocytes/spleen (combination 6 and 7).

Class II disparate reactions yielded varying numbers of lymphocytes. Those combinations which were directed against the I-Ak molecule yielded the most cells/spleen (combination 8 and 9) 2.8 to 4.1 x 106, while (B10.GD x








Table 4.
Summary of the number of recovered cells obtained fran the GVH spleens on day 5 of the GVH reaction. (a)

Genetic Cell number Number of Recovery of Canbination Disparity recovered mice used cells/spleen


1) B1O.BR anti-(B10xB10.BR)F H-2b 40.0 x 106 8 5.0 x 106 2) B6 anti-BALB/c H-2d 56.5 x 106 11 5.1 x 106 3) (Bl0xBl0.Q)F1 anti-BlO.BR H-2 27.3 x 10 11 2.5 x 10

4) (B10.A(4R)xB10.GD)F1
anti-Bl0 KbI-Ab 14.6 x 106 10 1.5 x 106 5) B1O.S(9R) anti(B10.HTTxB10.A)F1 KI-Ak 11.0 x 106 10 1.1 x 106

6) Bl0.MBR anti-(A.TLxB10.MBR)F1 KsDd 3.2 x 106 7 4.6 x 105

7) B0O.M(17R) anti-A/J Dd 2.2 x 106 10 2.2 x 105

8) B10.S(9R) anti(Bl0.HTTxB10.TL)F I-Ak 41.0 x 106 10 4.1 x 106 9) (Bl0xBl0.Q)F anti-Bl.MR I-Ak 30.7 x 106 11 2.8 x 106 10)(B10.GDxBl0.Z)F1 anti-BlO I-A 4.8 x 106 10 4.8 x 105 11)B10.AQR anti(B1.T(6R)xBl0.AQR)F1 I-A 4.3 x 106 6 7.2 x 105

12)Bl0.A(4R) anti-BlO.A(2R) I-Ek 2.5 x 106 4 6.3 x 105

a) Host animals were sublethally irradiated and reconstituted with 40 x 10 donor
splenocytes. The animals were sacrificed on day 5. The spleens were collected and mrde into a single cell suspension and cultured overnight. The recovered viable cells were then counted and utilized in PLTs the next day.

IQ








63



Bl0.MBR)F1 anti-Bl0 (combination 10) gave the least, 4.8 x 105 cells/spleen. Thus, it appears that I region GVH reactions are the most variable in terms of obtaining viable lymphocytes from the fifth day of the GVH reaction.


J. Functional activities of leukocytes obtained from GVH animals


The lymphocytes recovered from these GVH affected

tissues were then tested in functional tests to determine what antigen(s) the recovered lymphocytes are capable of reacting. Two functional tests have been used: 1) the mixed lymphocyte reaction/primed lymphocyte typing test which measures the ability of reactive cells to proliferate against foreign histocompatibility antigens, 2) the cell mediated lympholysis test which assays the ability of primed cells to lyse 51Cr labeled target cells possessing the appropriate antigens.


i. Mixed lymphocyte reaction/primed lymphocyte test. B10.BR anti-(B10 x B10.BR)F1 reaction

The spleen cells recovered from the (B10 x B10.BR)F1 mice undergoing GVHD induced by B1O.BR cells were found capable of proliferating in a mixed lymphocyte reaction. The kinetic responses of these lymphocytes are presented in Figure 19. The activated cells reacted to B10 (H-2b























Figure 19. The proliferative responses of GVH primed splenocytes derived from an acute GVHR: B10.BR anti-(Bl0 x BIO.BR)F1. This figure demonstrates the kinetic responses of GVH primed cells obtained fran the spleens of (B10 x BlO.BR)F mice reconstituted with B1O.BR cells on day 10. The primed cells were cultured with eit r: BlO.BR, C57BL/q or Bl0.D2 irradiated stimulator cells. The various cultures were then pulsed with 1 pCi of H-TdR for 12 hrs on days 1 to 5.







65









0 C\j W 0
O

C13


Ec


- ----- L





o










0 00
OOOO




01 dr o o o o I o 0 0 0 0 0 0n 0, 03 00 0 0 0 0








66



cells, expressing antigens of the F1 host to which the BlO.BR lymphocytes were sensitized. No proliferative activity was directed against the syngeneic donor cells which initiated the GVH, namely B10.BR (H-2k). This indicates that the host's lymphocytes are not proliferating against the donor lymphocytes, as in the nonirradiated F1 animals (see Table 3). Nor is reactivity seen against B10.D2, an unrelated third party haplotype (H-2d). Some cross reactivity of the response of the primed cells is seen on day 1, but this is not considered significant because of the loss of activity on day 2. The optimal peak of secondary 3H-TdR incorporation of either MLR or GVH primed cells is always found on day 2. Thus, the cells do not seem to be capable of reacting towards antigens which are different from those antigens which initially triggered the GVHR; and this makes this reaction appear to be similiar to the in vitro primed lymphocyte test (PLT). In addition, GVH activated lymphocytes do not react towards the T cell mitogens, concanavalin A (Con A) and phytohemagglutinin (PHA), or lipopolysaccharide (LPS), a potent murine B cell mitogen (Table 5).

(B10 x BlO.BR)F1 hybrid mice reconstituted with

parental B10.BR spleen cells normally died between days 10 to 14. Cell populations recovered from the spleens and livers of mice showing signs of severe GVHD on day 8 and 9








67








Table 5.
Inability of GVH primed cells to respond to mitogens.


3H-TdR incorporation CPM+SD (a).


addition normal GVH normal cells normal cells
cells primed treated with treated with
(b) cells anti-Thy+C' anti-I-A+C'
(c) (d) (e)


none 4668+880 1349+12 6303+ 1477+ 163

Con A(f) 54043+6244 417+94 2700+595 22311+1345

LPS (g) 98977+5401 2531+81 71110+655 19737+ 517



a) MLR assays xere harvested at 48 hrs following a 12 hr pulse
with 1 pCi H-TdR.
b) Spleen cells from a normal healthy mouse (BIO.BR) were used
in this study. 5 x 10 cells were placed in each well of a
microtiter plate.
c) (B10 x B1O.BR)F mice were sublethally irradiated and
reconstituted with 40 x 106 B1O.AQR splenocytes. The mice
were sacrificed on day 5 after reconstitution.
d) Normal splenocytes were treated with anti-Thy 1.2 antibody and
complement immediately before culturing the cklls.
e) Normal splenocytes were treated with anti-I-A antibody and
complement immediately before culturing the cells. f) Ccncanavalin A dose was 5 pg/ml. g) Lipopolysaccharide dose was 100 pg/ml.








68




after engraftment were compared in PLT with in vitro activated B1O.BR anti-(Bl0 x B1O.BR)F1 PLT cells. As presented in Table 6, major differences exist in the response pattern of the in vivo and in vitro activated cell populations. Cells obtained from either the spleens of livers of the F1 hybrids undergoing GVH were restimulated by cells from strains carrying the H-2b haplotype. However, stimulation with cells from recombinant strains possessing either the H-2Db or H-2Kb antigens, e.g. B10.MBR, BlO.A(2R) produced little if any reactivation of proliferation, and no cross reactivity was observed on unrelated third party strains. In contrast, MLR generated PLT cells exhibited strong secondary responses to strains carrying the H-2b haplotype as well as H-2b haplotype derived K and D region antigens. In addition, cross reactivity on unrelated third party strains was observed such as B1O.T(6R).

While a number of factors may account for the different patterns of reactivity exhibited by in vivo and in vitro primed cells, the most likely one is the protocol used herein to obtain responding PLT cell populations from organs undergoing GVH reactivity. Overnight incubations of the in vivo primed cells in the presence of cell debris and dying PMNs could alter markedly the reactive patterns. To determine the effect of this procedure, cell mixing







Table 6.
Comparison of PLT using BIO.BR anti-(B10 x BO.BR)F1 cells generated either in MLR or in GVH.


3H-TdR Incorporation CPM+SD (a)

H-2 Genetics
Stimulator In vitro primed (b) In vivo primed (c) strain KA JED spleen liver


B10.BR k k k k k 7344+ 233 1429+ 252 234+127 (Bl0xBl0.BR) b b b b b
F1 k k k k k 92117+ 3026 44871+ 402 4015+984 B10 b b b b b 83972+ 4 42062+1809 4467+209 B10.A(2R) k k k k b 19769+ 1744 3462+1744 278+ 83 Bl0.MR b k k k q 50799+13119 2929+ 11 189+ 35 B10.A(4R) k k b b b 25089+ 204 2067+ 264 111+ 40 Bl0.GD d d b b b 49169+ 1409 7191+1222 319+ 22 B10.A(3R) b b b k d 68108+ 1971 36910+5474 3553+458 B10.A(5R) b b k k d 87682+ 2052 34182+1667 1653+308 Bl0.T(6R) q q q q d 40085+ 2713 6146+ 293 412+24 a) PLTs were harvested at 48 hrs following a 12 hr pulse with 1 pCi of 3H-TdR. b) Primary MLR consisted of B10.BR splenocytes incubated with irradiated (B10 x
BIO.BR)F splenocytes for 8 days.
c) GVH reactivity was induced in sublethally irradiated (BlO x B1O.BR)F1 mice with
BIO.BR splenocytes. On day 8, the spleens and livers were collected, made into
single cell suspensions and cultured overnight prior to use.








70



experiments were performed. Responding cell populations primed in the MLR of BlO.BR responding against (B10 x B10.BR)F1 were mixed with equal numbers of cells removed from the spleens of (B10 x B10.BR)F1 mice undergoing GVH reactivity following reconstitution with B10.BR spleen cells. Following an overnight incubation, this mixed cell population was tested in PLT and its reactivity compared to the responses of the two individual cell populations. As shown in Table 7, Column 3 the mixed cell population exhibited the reactivity of the in vitro primed cell population (Column 1).

In a more direct approach to determine if differences arise due to culture artifacts, cells obtained from spleens undergoing severe GVHD separated on Ficoll-Hypaque density gradients (d=1.101) were examined directly. This procedure results in a responding purified population contaminated with less than 2% PMNs. As shown in Table 7, Column 4, the responding cell population obtained in this manner still exhibited a PLT reactivity identical to the in vivo primed cells which had been incubated overnight. Thus, the differences in the reactivity of the in vivo versus in vitro primed cells does not appear to be dependent on the handling of the cells.

K and D antigens are known to exist on every cell of the body, while I region antigens are found on only a







Table 7.
GVH primed cells fail to alter MER primed cell responses.



3H-TdR Incorporation CPM+SD (a)

H-2 Genetics In vitro In vivo In vivo + Purified Stimulator primed primed In vitro in vivo Strains K A J E D cells (b) cells (c) cells (d) cells (e)


BIO.BR k k k k k 10144+2758 1251+ 12 11820+1602 1541+ 387

B10 b b b b b 97862+2944 26131+1610 83250+1463 26037+ 380 BlO.A(3R) b b b k d 82811+2935 23096+ 373 73750+4613 35933+2444 B10.A(5R) b b k k d 91105+1805 20935+ 499 73345+ 516 39068+ 176 Bl0.MBR b k k k q 45266+ 512 3858+ 564 31563+3743 6270+ 201 BlO.A(4R) k k b b b 28264+1404 2692+ 443 26423+1596 2390+ 160 Bl0.WB j j j j b 33018+3188 3940+ 330 29996+3005 8278+2161


a) PLTs were harvested at 48 hrs following a 12 hr pulse with 1 pCi of 3H-TdR.
b) Primary MLR consisted of B1O.BR splenocytes incubated with irradiated B10
splenocytes for 6 days.
c) GH reactivity was induced in sublethally irradiated (B10 x B10.BR)F mice with
BlO.BR splenocytes. On day 5, the spleens were collected, made into a single cell
suspension and cultured overnight prior to use.
d) Equal numbers of in vitro and in vivo primed cells were cultured together
overnight prior to use.
e) Single cell suspensions of GVH splenocytes were centrifuged over a Ficoll-hypaque
density gradient (d=1.101). The resulting cells contained <2% PMNs. After a single
wash, the resulting lymphocyte population was tested in PLT.








72



selected set of cells such as B cells, monocytes, skin dendritic cells and liver Kupffer cells. The possibility exists that the cells which react to the Kb and Db antigens are filtered out before they home to the spleen or liver. To exclude this possibility, embryonic B6 mouse fibroblast monolayers were established and passaged three times. These cells possess the Kb and Db antigens, and do not express the I-Ab antigen. In vitro primed B10.BR anti-(Bl0 x B1O.BR)F1 cells were then incubated for 2 hours on this monolayer and then gently rocked off the monolayer. Approximately one half of the primed cells were removed by this treatment. These cells as well as a sample of the original primed population were tested in PLT. The results shown in Table 8 show that both populations were still capable of responding to the Kb and Db antigens found on B10.MBR and B10.GD. Although cytotoxic T cell activity was not tested before and after adsorption, it appears that this filtering mechanism does not occur in vivo because cytotoxic T cells are found in vivo in the spleen and liver and that they do respond to the Kb and Db antigens (see below).

Treatment of the in vivo primed B1O.BR anti-(Bl0 x

BIO.BR)F1 cell populations with anti-Thy 1.2 antibody plus complement totally abolishes the PLT reactivity (Table 9). Treatment with monoclonal anti-I-Ak antibody plus








73








Table 8.
Absorption of MLR primed lymphocytes fails to remove
proliferative responses towards K/D antigens.



3H-TdR Incorporation CPM+SD (a)

H-2 Genetics
Stimulator Treatment: (b)
strain K A J E D none absorption


none 8236+1126 8478+2284 BIO.BR k k k k k 11043+1766 9187+ 184 B6 b b b b b 88425+ 648 69299+6033 Bl0.A(3R) b b b k d 99479+3454 83931+1179 B10 .MER b k k k q 15213+2087 10245+2835 Bl0.GD d d b b b 52650+ 47223+6376 B10.A(4R) k k b b b 15215+ 42 15057+1344



a) PLTs we e harvested at 48 hrs following a 12 hr pulse with 1
pCi of H-TdR.
b) Primary MLR consisted of B10.BR splenocytes incubated with B10
splenocytes for 5 days. An aliqout of primed cells was
incubated for 2 hrs, while the remaining cells were incubated
over the B6 fibroblast monolayer for 2 hrs. After the
incubation the cells were gently rocked franom the plate and the lynphoblasts were recovered. The fibroblast monolayer was made
by taking B6 emnbryoes and mincing then into single cell
suspensions using a 0.5% trypsin solution. The fibroblast
monolayer was passaged three times during a period of 18 days.








74






Table 9.
The PLT activity of GVH primed cells after treatment with various
monoclonal antibodies plus complement.



3H-TdR incorporation CPM +SD (a)


stimulator primed cells primed cells primed cells cells treated with treated with treated with
C' (b) anti-Thy + C' anti-I-A + C'
(c) (d)


none 604+382 1574+544 1778+266 B10 6409+175 1490+ 37 16711+880 BIO.BR 649+ 69 1762+ 50 2252+ 47 (Bl0xBl0.BR)F1 6361+239 1701+ 33 17570+544 B10.A(3R) 5478+ 59 1619+ 67 14862+145 B0 .A(4R) 623+ 45 1152+183 3360+771 BIO.A(5R) 3320+494 1128+ 79 13892+951



a) GVH reactivity was induced in sublethally irradiated (Bl0 x
Bl0.HR)F1 mice with B10.HR splenocytes. Eight days later the
spleens of these animals were removed and prepared into a
single well suspension and used. The cultures, consisting of
30 x 10 cells per well,were pulsed with 1 pCi of H-TdR
for 8 hrs on day 2 of the reaction.
b) Spleen cells recovered from the GVH reaction were treated with
rabbit complement alone.
c) Spleen cells recovered from the GVH reaction were treated with
anti-Thy antibody plus complement immediately before
culturing.
d) Spleen ckls recovered from the GVH reaction were treated with
anti-I-A antibody plus complement immediately before
culturing.








75



complement (kills 60% B10.BR splenocytes and 5% B10 splenocytes) a procedure which enriches for T lymphocytes by depleting the 30 to 40% contaminating B lymphocytes, increased responsiveness in those combinations already exhibiting a positive response, but did not produce reactivity against strains which showed negative responses.



C57BL6 anti-BALB/c reaction

Irradiated BALB/c mice reconstituted with B6 spleen cells normally died by day 9 to 10 after engraftment. The pathology was identical to that found in the B1O.BR anti(B10 x B10.BR)F1 combination. Cell populations recovered from the spleens, livers and lymph nodes on day 6 were tested in PLT for their ability to mount a secondary proliferative response. As can be seen in Table 10, all three populations exhibited strong proliferative responses against cells from mice possessing the H-2d haplotype, e.g. BALB/c and DBA/2 or from mice possessing the H-2Kd and I-Ad regions, e.g. B10.GD. Little, if any significant reactivity was elicited against third party strains including those expressing H-2Dd region antigens, e.g. B10.A(3R), Bl0.A(5R) and BlO.T(6R). Thus, these alloactivated cells appear to recognize primarily determinants encoded by genes located on the left side of the H-2 complex.








Table 10.
Comparison of PLT of B6 anti-BALB/c cells generated either in MLR or in GVH.



3H-TdR Incorporation CPM+SD (a)
H-2 Genetics
Stimulator In vitro primed In vivo primed cells (c) Strains K A J E D cells (b) Spleen Liver Lymph nodes


B6 b b b b b 3058+1990 1203+ 71 1203+ 71 726+ 147

BALB/c d d d d d 70295+4777 23930+ 555 23522+2619 37980+2100 DBA/2 d d d d d 97073+1100 50130+1773 49024+2128 43727+2756 Bl0.GD d d b b b 55067+6394 26637+1414 20612+2598 39550+2308 Bl0.A(3R) b b b k d 22429+1108 2847+ 710 3216+ 954 B10.A(5R) b b k k d 24162+9856 1551+ 909 376T 215 4068+ 199 Bl0.T(6R) q q q q d 22962+ 953 2573+ 984 1398+ 469 7801+ 995

Bl0.A(2R) k k k k b 18534+8060 2179+ 735 1273+ 78 247+ 66 BlO.A(4R) k k b b b 8139+ 492 798+ 217 1106+1044

WRq q q q q 13433+ 619 1904+ 85 592+ 28 AKR k k k k k 12294+1568 1855+ 327 522+ 40



a) PLTs were harvested at 48 hrs following a 12 hr pulse with 1 pCi of 3H-TdR. b) Primary MLR consisted of B6 splenocytes incubated with irradiated BALB/c splenocytes
for 7 days.
c) GVH reactivity was induced in sublethally irradiated BALB/c mice with B6
splenocytes. On day 6, the spleens were collected, made into a single cell
suspension and cultured overnight prior to use.








77



Secondary responses of B6 anti-BALB/c PLT primed cells primed in MLR are also included in Table 10 for comparison. These cells exhibited a quite different pattern of reactivity from GVH primed cells. Strongest responses were elicited against strains possessing the H-2d haplotype or the H-2Kd and I-Ad regions; however, strong responses also occurred against strains possessing the H-2Dd region antigens. In addition, cross reactivity against a number of unrelated third party strains, e.g. BlO.A(2R), SWR and AKR, was observed.



The Bl0.RIII anti-BlO.A(5R) reaction

In another whole H-2 disparate GVH, B10.RIII antiBlO.A(5R), similar results were found (Table 11) like those in the two previous combinations. The MLR generated primed cells respond towards the Kb and I-Ab antigens present on BlO.A(5R) and B6 mice. Strong stimulation was also found on mouse cells possessing the Kb antigen Bl0.MBR as well as the I-Ek and Dd antigens, e.g. B1O.A, Bl0.AQR, B10.TL, BlO.S(9R), A.TFR5, BlO.M(17R) and B10.BR. In contrast, GVH primed cells only recognized the BlO.A(5R) and B6 mouse cells. The lack of a response to the B10.MBR cells apparently eliminates stimulation due to the Kb antigen.







78




Table 11.
Comparison of PLT of BlO.RIII anti-Bl0O.A(5R) cells
generated either in MLR or in GVH.



3H-TdR Incorporation CPM+SD (a) H-2 Genetics
Stimulator In vitro primed In vivo primed Strains K A J E D cells (b) cells (c)


none - - 3832+ 2227 489+ 257 BlO.RIII r r r r r 4229+ 858 151+ 23 Bl0.A(5R) b b d d d 85395+ 1382 30006+ 168 B6 b b b b b 93669+ 1283 26426+4045 Bl0.D2 d d d d d 51594+ 7504 5886+1625 Bl0.AQR q k k k d 66655+ 2270 2873+1453 B10.A k k k k d 57325+ 49 1704+ B10.TL s k k k d 52795+15596 2814+1071 BlO.S(9R) s s k k d 28673+ 3896 468+ 251 A.TFR5 f f k d 32065+ 8188 374+ 173 Bl0.MBR b k k k q 43565+ 1493 2398+ 155 B0.M(17R) k k k k f 27360+ 2955 1533+1327 BIO.BR k k k k k 20357+ 3594 3608+ 91 Bl0.A(4R) k k b b b 19194+ 1996 610+ 1 BIO.GD d d b b b 21432+ 2476 997+ 250 B10.WB j j j j b 8476+ 888+ 347 Bl0.M f f f f f 11401+ 2331 1148+ 286



a) PLTs wese harvested at 48 hrs following a 12 hr pulse with 1
pCi of H-TdR.
b) Primary MLR consisted of BlO.RIII splenocytes incubated with
irradiated Bl0.A(5R) splenocytes for 6 days.
c) GVH reactivity was induced in sublethally irradiated Bl0.A(5R)
mice with B1O.RIII splenocytes. On day 5, the spleens were collected, made into a single cell suspension and cultured
overnight prior to use.








79




Again, the GVH primed cells display a more restricted response pattern than do primed cells established in vitro.

The B10.GD anti-Bl0O.M(17R) and B1O.WB anti-Bl0.M(17R) reactions

Two other genetic combinations with H-2 disparate

regions studied were Bl0.GD anti-Bl0O.M(17R) and Bl0.WB antiBlO.M(17R) as shown in Table 12. Both in vitro primed cell populations responded towards cells from mice which possessed either the Kk,I-Ak or Df molecules. In contrast, the GVH primed cells reacted only to those cells which possessed the I-Ak molecule, e.g. B1O.BR, BlO.M(17R), Bl0.MBR, B10.AQR and BlO.A(4R). Although minor reactivity is seen against Ek and D molecules, the GVH primed cells clearly did not respond to the same magnitude as the in vitro primed cells do.

Even though the B10.GD anti-Bl0O.M(17R) and B1O.WB antiBl0.M(17R) reactions are directed against the B1O.M(17R) haplotype, these two in vitro reactions are not identical. The Bl0.GD anti-BlO.M(17R) primed cells recognized some determinants which the Bl0.WB anti-Bl0O.M(17R) primed cells did not, such as those found on the B1O.F cells. What these determinants are is not readily distinguished from the present data. However, this antigenic determinant is readily detected by the in vivo primed cells giving some







Table 12.
Comparison of PLT of anti-Bl0O.M(17R) reactive cells generated either in MLT or in GVH.



3H-TdR Incorporation CPM+SD (a)

H-2 Genetics In vitro primed (b) In vivo primed (c)
Stimulator BIO.GD anti- BlO.WB anti- B1O.GD anti- B1O.WB antiStrains K A J E D Bl0.M(17R) B10.M(17R) Bl0.M(17R) Bl0.M(17R)


none 1594+ 281 1552+ 638 97+ 55 579+ 23 B10.GD d d b b b 1115+ 367 4062+1191 2998+ 0 968+ 303 B10.D2 d d d d d 2539+ 199 1561+ 194 2864+ 317 BI0.WB j j j j b 2093+ 506 614+ 146 Bl0.M(17R) k k k k f 44694+ 1549 24304+1946 34989+2594 28647+ 984 B10.BR k k k k k 55622+ 1332 26465+ 271 39904+3731 36015+ 921 B10.MBR b k k k q 44515+12015 24192+2344 36908+ 0 35239+1383 B10.AQR q k k k d 39051+ 2837 43982+2751 36945+4965 35172+3935 Bl0.A(4R) k k b b b 18540+14060 34730+3325 27239+3950 21024+5081 B1O.S(9R) s s k k d 17084+ 799 4483+ 757 7780+ 598 1620+ 506 B1O.M f f f f f 34894+ 12080+1502 9843+ 300 3645+ 376 B1O.RIII r r r r r 4489+ 134 2164+ 158 BO.S s s s s s 353+ 341 4462+ 366 BIO.F p p p p p 21093+ 184 9035+ 833 12240+1264 3509+ 290 B10.T(6R) q q q q d 9027+ 904 5535+ 306


a) PLTs were harvested at 48 hrs after a 12 hr pulse with 1 VCi of 3H-TdR. b) Primary MLR consisted of either BlO.GD or BlO.WB splenocytes incubated with
irradiated B1O.M(17R) splenocytes for 6 days.
c) GVH reactivity was induced in sublethally irradiated B10.M(17R) mice with either
Bl0.GD or BlO.WB splenocytes. On day 5, the spleens were collected, made into single
cell suspensions and cultured overnight prior to use. Co
O








81



possibility that these cross reactions could either be due to cross reactivity to the I-AP molecule which is common to the I-Ak molecule or that the I-Ep cross reacts with the I-Ek molecule. The latter explanation is favored because the Bl0.GD does not express the I-E molecule and is therefore capable of recognizing the entire I-E molecule as foreign, while the B1O.WB animal does express the I-E molecule and does not have to respond to an entire I-E molecule as would the B10.GD cells.



The B1O.M(17R) anti-Bl0.GD and B1O.WB anti-BlO.GD reactions

When another set of whole H-2 reactions is established against the B10.GD mouse [B10.M(17R) anti-Bl0.GD and B10.WB anti-Bl0O.GD] a different pattern of data is seen (Table 13). The predominate in vitro stimulatory antigens come from the left side of the H-2 complex, i.e. Kd and I-Ad as exhibited by Bl0.GD and B10.D2. Reactivity is also observed on B6, BlO.A(3R) and B10.A(4R) cells which possess the I-Jb and Db antigens. In addition, cross reactivity on the I-Ab molecule is also postulated for the B6 and BlO.A(3R) cells, but these responses to the antigens do not seem to be additive when compared to the B1O.A(4R) response. This cross reactivity on I-Ab could be due to the shared antigens Ia8 and Ial5 which I-Ad has in common with I-Ab to which the I-Ak molecule lacks. In addition, the GVH






Table 13.
Comparison of PLT of anti-B10.GD reactive cells generated either in MLR or in GVH.


3H-TdR Incorporation CPM+SD (a)

H-2 Genetics In vitro primed (b) In vivo primed (c)
Stimulator Bl0.M(17R) B10.WB B1O.M(17R) B10.WB Strains K A J E D anti-Bl0.GD anti-Bl0.GD anti-Bl0.GD anti-Bl0.GD


none 355+ 247 2699+ 19 262+ 245 417+226 Bl0.WB j j j j b 2243+ 286 396+146 BIO.M(17R) k k k k f 1910+ 83 4204+ 743 241+ 2 350+120 B10.GD d d b b b 56739+1411 30192+1219 34250+4848 9604+556 B10.D2 d d d d d 67771+7652 33813T 197 38583T 921 19700+851 Bl0.A(3R) b b b d d 26543+2309 9305+ 193 11502+ 765 1912+253 B6 b b b b b 25131+ 14670+1843 11175+ 2035+ 78 B10.A(4R) k k b b b 41841T 861 6302+ 137 1278+ 83 869+ 47 BI0.BR k k k k k 5766+1809 4062+1102 186+ 17 652+227 BIO.MR b k k k q 4720+1587 4981+ 90 239+ 34 476+218 BIO.AQR q k k k d 2612+3271 6343+ 689 209+ 38 540+T 10 B1O.HTT s s s k d 12072+2399 2025+1145 BIO.TL s k k k d 14107+1206 4762+ 838 537+ 141 439+199 B10.S(9R) s s k k d 13030+ 436 4309+ 46 2053+ 368 326+ 74 B1O.S s s s s s 13043+2728 2811+1221 B1O.M f f f f f 1203+ 214 4775+ 687 2091+ 471 1071+ 52 B10.Q q q q q q 34612+3890 10860+ 921 B1O.RIII r r r r r 4519+ 550 452+ 1


a) PLTs were harvested at 48 hrs following a 12 hr pulse with 1 pCi of 3H-TdR. b) Primary ER consisted of either B10.M(17R) or BlO.WB splenocytes incubated with
irradiated B1O.WB splenocytes for 6 days.
c) GVH reactivity was induced in sublethally irradiated B1O.GD mice with either
B10.M(17R) or B1O.WB splenocytes. On day 5, the spleens were collected, made into
single cell suspensions and cultured overnight prior to use.








83




primed cells also appear to recognize these antigens, since GVH primed cells apparently only respond to I region antigens. Thus, this cross reactivity is likely to be due to class I antigens.

Interestingly, the in vitro Bl0.M(17R) anti-BlO.GD reaction appears to detect a cross reaction on the Ks molecule, since Bl0.TL, B10.S, BlO.S(9R) and B10.HTT cells induce stimulation by the primed cells. In contrast, the GVH primed cells do not appear to recognize this cross reactivity; again the response seen by the GVH primed cells is different from those of the MLR generated cells. These cross reactions to Ks which are seen by the in vitro primed BlO.M(17R) anti-Bl0O.GD cells are not seen by the in vitro primed Bl0.WB anti-BlO.GD cells.



The I region GVH reactions

A number of mouse strain combinations exists with limited I region disparity. Two of the most extensively studied combinations are BlO.A(4R) responding against BlO.A(2R), two congenic lines differing genetically between the H-2 I-A and D, and BlO.AQR responding against BlO.T(6R), two congenic lines differing genetically throughout the I regions. These two combinations are considered to represent, respectively, an anti-I-Ek and anti-I-Aq response. The secondary MLR responses of B1O.A(4R) anti-Bl0.A(2R) and







84




Bl0.AQR anti-(Bl0O.T(6R) x B1O.AQR)F1 PLT obtained from 10 day GVH spleens or primary MLR are examined in Table 14.

As presented in Table 14, the PLT cells generated in MLR exhibited strong responses against the primary stimulating strains as well as strong cross reactivity on third party strains. For example, in the B10.A(4R) antiBlO.A(2R) reaction the specific antigen is I-Eak k,yet ks
B10.HTT which has I-E 8 is recognized by the in vitro primed cells as well as the BlO.RIII cells. Likewise, a great deal of cross reactivity is seen by the B10.AQR anti-(BlO.T(6R) x B10.AQR)F1 reaction, a reaction supposedly directed at the I-Aq molecule. Mouse cells from B10, BlO.S(7R), BlO.M and BlO.RIII mice elicit a response by these primed cells. Again as was demonstrated with the other reactions the GVH primed cells demonstrate marked specificity. In the B10.A(4R) anti-Bl0O.A(2R) reaction the GVH primed cells only respond to the specific antigen: I-E akk found on Bl0.AQR, BlO.A(2R) and BlO.A. While in the B10.AQR anti-Bl0.T(6R) reaction only those cells possessing the I-Aq molecule found on B1O.T(6R) and DBA/l cause restimulation.



K/D region GVH reactions

Two strains namely Bl0.MBR and A.TL possess genetic

differences at only K and D loci. The MLR and GVH reactivity




Full Text

PAGE 1

THE IMMUNOBIOLOGY OF GRAFT VERSUS HOST DISEASE AND ITS ATTEMPTED PREVENTION USING NATURALLY OCCURRING SUPPRESSOR FACTORS BY MARTIN ROBERT JADUS A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1983

PAGE 2

ACKNOWLEDGEMENTS I would like to express my appreciation to the many people who helped me complete this dissertation. To my parents, without whose help and understanding I would never have made it this far, I extend my gratitude. I would like to express my deep thanks to Dr. Ammon Peck for his help and guidance through this work, especially his willingness to give me a free hand performing these experiments, related or unrelated to this dissertation which allowed me to satisfy my insatiable curiosity. In addition, his friendship is warmly remembered. Appreciation is also given to Drs. Ward Wakeland and Art Kimura for their discussions and insights about the immunogenetics of the mouse system. Additional thanks are given to Dr. Wakeland and Vicki Henson for their preparation of monoclonal antibodies and Dr. Kimura for the column chromatography Thanks are also given to Dr. Paul Klein for whom a generous supply of materials allowed me to get through the first year of this project and Drs. Richard Smith and Shiro Shimuzu from whom I was able to acquire enough interleukin 2 in order to maintain and expand the cytotoxic T cells which were important to this study. Thanks are also extended to Dr. Micheal Norcross for helping with the photography. ii

PAGE 3

Thanks are also extended to Drs. Roy Weiner and Noel McClaren who shared their help and time while participating on my advisory committee. Another round of appreciation is given to those people who helped with the mice: Mr. Lucetta for breeding the necessary mice for most of this work, Dr. Bobby Collins for his preparation of the histology of the various tissues, as well as Steve Noga for his help in interpreting the various histological and pathological sections which were reported. Thanks are also given to Dan Cook and Dr. Jian Xiang Huang who helped me with the word processing of this work. Without their help this epic manuscript would have cost a fortune to produce. A special round of gratitude is given to the excellent technical help of Laura Prall, Jean Mahlman and Mary Ann Searle who helped me with the arduous and time consuming tasks such as preparations of various materials which allowed me the necessary time to plan further GVH conquests. Thanks also go to Vi Sudipong for her participation with the minor histocompatibility work during the summer of 1982. Finally, a word of gratitude is given to my fellow graduate students: Vicki Henson, Ted Hall, Kim Peeler and Dave Shaut who had to share in my moods when things were not correct. Ill

PAGE 4

TABLE OF CDNTElSrrS page ACKNOWLEDGEMENTS ii LIST CF TABLES vii LIST CP FIGURES xi PiBSTRACI xiv INTRODQCTION 1.1 Tte Graft Versus Host Reaction 1 A. Sytemic 5 B Local 6 1.2 Tte In Vitro Measurement of the GVH Reaction 9 1.3 Controlling the Activities of T Cells In Vitro 20 1.4 Svppressor Cells 21 1.5 Tte Rationale for Ttese Experiments 23 MATERIALS AND MBTHCDS 2 1 Animals 24 2 2 Antisera 25 2.3 Canplement Dependent Antibody Cytotoxicity 26 2.4 Fluorescent Microscopic Determination of Cells 26 2.5 Cell Preparations 27 2.6 Primary Mixed Lymphocyte Reaction 27 2.7 Induction of Graft Versus Host Reactivity 28 2.8 Primed Lymphocyte Typing Tests 28 2.9 Cell Mediated Lympholysis Assays 29 2.10 Preparation of Purified Interleukin 2 30 2.11 Preparation of Newborn Spleen Cells 31 2.12 Preparation of Alpha Fetoprotein 32 RESULTS 3.1 Inability of Allogeneic Cells to Induce GVHD in Normal Adult F, Mice 33 3.2 Generation of GJBD in Adult F, Mice Immunosuppressed Through Irradiation 34 A Study of the survival rates of mice lethally or sublethally irradiated 34 B Sublethally irradiated host provide an environment for GVHD 36 IV

PAGE 5

OONTENTS— Continued C Histology of sublethally irradiated (650r) mice 40 D Pathology of GVHD in MHC disparate strains 40 E The effect of the host's age upon generation of GVHD 41 F Histopathological examination of v*iole H-2 disparate GJED revealed marked effects 46 G Pathology of K/D or I region disparate GVHD 53 H Cellular composition of the host organs undergoing GVHD 60 I Recovery of viable lymphocytes fran various histoincompatible differences 51 J Functional activities of leukocytes obtained from GVH animals i mixed lyirphocyte reaction/primed lynphocyte best 63 ii cell mediated lympholysis assays 96 K Ability of primed lytiphocytes to cause mortality in sublethally irradiated mice 136 L Histology and pathology of secondary disease 141 M Mortality induced by anti-I-A^^ long term cultured T cell lines and clones 150 3.3 Attempts to Modify GVHD A Attefnpts to prevent lethal GVHD using concentrated monoclonal antibody directed towards the host's I-A molecule 15 4 B Attempts to prevent lethal GVHD using neonatal splenocytes i. Inhibition of acute lethal GVHD using CBA/J newborn spleen cells 157 ii Histopathology of the experimental and control host animals 15 8 iii. Functional reactivity of donor cell populations after initial period of culturing 161 iv. Functional reactivity of ly[tphocytes derived from long term surviving host mice 167 V. Genetic restrictions in the ability of CBA/J newborn cells bo suppress lethall GVHD 175 vi. The presence of newborn spleen cells incapable of suppressing the GVH reactivity of adult cells fails to modify the response of the sensitized donor cells 178 V

PAGE 6

OONTENTS— Continued C Attempts to prevent GVHD using newborn spleen supemates i. Characterization of the newborn supemate — 178 ii. Size profile of the newborn supemate factors 180 iii. Time of addition studies of the newborn factors 190 iv. Effects of the si:5)emate in GVH 193 D Attenpts to prevent lethal GVHD using AFP 196 Discussion 4.1 The N^d for Inmunosuppression for the Development of G7HD 201 4.2 The Kinetics of GVHD 202 4.3 Tte Functional Activities of the T lymf^ocytes Recovered from GVHD Animals 203 4.4 The Ncnspecific Factors Influencing Mortality in GVHD 216 4.5 The Attempts to Prevent GVHD Using Anti-Hcst I-A Antibody 221 4.6 Tte Attempts to Prevent GVHD using Newborn Sippressor Cells 222 REFERENCES 235 BIOGRAPHICAL SKETCH 244 VI

PAGE 7

LIST OF TABLES Table page 1. Strength of MLR and GVHR across various regions of the H-2. 15 2. The effect of the route of injection in order to generate primed lyir^^ocytes in the BIO. BR anti-(B10 x B10.BR)F, reaction 35 3. Tte effect of radiation on the host in order to generate specific primed lympiiocytes in the BIO. BR anti-(B10 x BIO .BR)F, reaction. 39 4. Sunnary of the number of recovered cells ctotained frcni the G7H ^leens on day 5 of the reaction. 62 5. Inability of GVH primed cells to respond to mitogens. 57 6. ComEarison of PLT using BIO. BR anti-(B10 x B10.BR)F, cells generated either in MLR or in OJH. 69 7. 07E primed cells fail to alter MLR primed cell responses. 71 8. Absorption of MLR primed lymphocytes fails to remove proliferative responses towards K/D antigens. 73 9. Tte PLT activity of GJH pricied cells after treatitent with various monoclonal antibodies plus oarplement. 74 10. CanEHrison of PLT of B6 anti-BALB/c cells generated either in MLR or in C5/H. 76 11. Cctiparison of PLT of BlO.RIII anti-BlO.A(5R) cells generated either in MLR or in GVH. 78 12. Canparison of PLT of anti-BlO.M(17R) reactive cells generated either in MLR or in O/H. -80 13. Ccrtparison of PLT of anti-BlO.GD reactive cells generated either in MLR or in GJE. 82 14. Comparison of PLT anti-H-2I region reactive cells generated either in MLR or in (Ml. 85 15. Cotcarison of PLT of BIO.MBR anti-(A.TL x B10.MBR)F, cells ^nerated either in MLR or in GVH. i 87 vii

PAGE 8

lABLES— Continued 16. Catparison of PLT of BALB/c anti-IBA/2 cells generated eitter in MLR or in C5/H. 89 17. Canparison of PLT responses of CBA/Ca anti-AKR cells generated either in MLR or in OJE. 91 18. PLT responses of anti-DBA/2 primed cells generated in MLR. 92 19. Suitnary of minor histocompatibility in vitro assays in inbred H-2 mice 94 20. CML reactivity of prined BIO. BR anti-(B10 x B10.BR)F, oells generated either in MLR or in GVH. 97 21. CML reactivity of primed B6 anti-BALB/c cells generated either in MLR or in (57H. 99 22. CML reactivity of priited BIO.MBR anti-(A.TL x B10.MBR)F, oells generated either in MLR or in C3/H. 101 23. CML reactivity of GTE primed BIO.MBR anti-(A.TL x B10.MBR)F, supplemented in vivo with or without interleukin 2 10 6 24. CML reactivity of GTE primed B10.M(17R) anti-A/J cells supplanented in vivo with or without interleukin 2. 107 25. CML reactivity of BlO.AQR anti-(B10.T(6R) x B10.AQR)F, cells generated either in MLR or in (Ml. = 109 26. CML reactivity of GVH prirted BlO.AQR anti-(B10.T(6R) x B10.AQR)F^ oells expanded in vitro with interleukin 2. 110 27. Tte inability of moiXDclonal antibodies to block killing of BlO.AQR LPS blasts by GVH prined BlO.AQR anti-(B10.T(6R) x B10.AQR)F, cells expanded in vitro with interleukin 2. 3j^4 28. CML reactivity of G/H primed BlO.AQR anti-(B10.T(6R) x B10.AQR)Fj^ cells expanded with interleukin 2. 115 29. CML reactivity of G7H primed (BIO x B10.Q)F, anti-BlO.MBR oells expanded with interleukin 2. 121 30. CML reactivity of G7H prirted (BIO.MBR x B10.GD)F, anti-B6 cells expanded with interleukin 2. 125 Vlll

PAGE 9

'mBLES— Continued 31. Lack of CML reactivity of GfJH primed BlO.HTT anti-(BlO ,TL x BIO.HTDF^ cells expaivied with interleukin 2. 128 32. CML reactivity of primed BALB/c anti-DBA/2 cells generated in GVH. 132 33. CML reactivity of primed BALB/c anti-DBA/2 cells generated in GVH. 13 3 34. CML reactivity of primed CBA/Ca anti-AKR cells generated in GVH. 137 35. Suppression of lethal GVHD by newborn spleen cells in (BIO x Bl0.m)F, hosts reconstituted with saxii-allogeneic adult BIO. BR cells. 159 36. Suppression of lethal GVHD by ne^/bom spleen cells in B6 host animals reconstituted with allogeneic adult BIO. BR cells— 160 37. The ability of BIO. BR splenocytes incubated with CBA/J newborn splenocytes to respond to mitogens. 166 38. The inability of lymphocytes obtained from newborn CBA/J suppressed GVHD to respond in a PLT. 168 39. Responsiveness of splenocytes frao CBA/J newborn mediated acute GVHD suppressed mice bo treact to various stimulants. 40. The inability of cells ctained from a newborn CBA/J suppressed GVHD to respond in a CML. 171 41. The genetic restrictions in the suppression of lethal GVHD by newborn spleen cells in host animals reconstituted with adult allogeneic spleen cells 176 42. The ability of lyiphocytes obtained from r^wborn SWR ^lenocytes supplemented GVHD to respond in a PLT. 179 43. The suppression of primary MLR by adding newborn supemate factors at different times. ''191 IX

PAGE 10

TaBTiES— Continued 44. The effect of various substances on a primary mixed lynphocyte reaction 19 2 45. lack of GVHD st5>pression in AFP treated (310 x B10.ER)F, mice = — 19 8 46. Inabibility of pregnant fatnales to be suppressed from acute GVHD. 200

PAGE 11

LIST CF FIGURES Figure page 1. Tte laws of transplantation. — 4 2. ffejor histoconpatibility ocmplex of tte nouse. 11 3. (5/HR across different histoconpatibility loci. 14 4. Effect of irradiation upon nouse survival. 38 5. Lethal GVHD across najor histoconpatibility loci. 43 6. The lack of a correlation of tte host's age to develop lethal O/HD in a najor histocompatibility mismatch: BIO .BR anti-BlO .WB. 45 7. Mouse liver fron either a normal animal or from a sublethally irradiated animal en day 10. 48 8. Normal mouse liver, a higher magnification. 48 9. The liver of a mouse undergoing acute GVHD. 50 10. A higher magnification of the previous liver. 50 11. The intestines of a normal mouse. 52 12. The intestines of a mouse undergoing acute GVHD. 52 13. Tte spleen from a normal mouse. 55 14. The normal spleen, a higher magnification. 55 15. Tte spleen of an animal undergoing GVHD. 57 16. A higher magnification from the previous tissue. 57 17. Further nagnif ication of the previous spleen. 59 18. A cytocentrifuge preparation of cells obtained from a nouse spleen undergoing acute G7HD. 59 19. Tte proliferative responses of Gm primed splenocytes cterived from an acute GVHR: BIO. BR anti-(B10 x B10.BR)F,. 65 20. The effect of purified IL 2 given to mice undergoing KA> GVHD: BIO.MBR anti-{A.TL X BlO.MBR)F.j^. 104 xi

PAGE 12

FIGURES — Continued 21. The effect of purified IL 2 given to mice undergoing I region GVHD: BIO.AQR anti-(B10.T(6R) x B10.ftQR)F,. 113 22. Lethal GVHD in sublethally irradiated (B10.T(6R) x BIO AQR )F^ 118 23. Lethal GVHD across ICI-A or I-A^ region differences. 120 24. Lethal GJHD in sublethally irradiated C57BL/10. 123 25. Lethal GVHD in k'^-a'^ or I-A^ mi snatches. ^^ 26. Effect of minor histocanpatibility antigens on survival in DBA/2 mice 130 27. Effect of minor histoccxnpatibility antigens on survival in AKR/J mice reconstituted with CBA/CaH cells. 135 28. Ability of 10^ (B10.A(4R) x B10.GD)F, anti-BlO priited cells to cause mortality in sublethally irradiated mice. — 139 29. The liver of a BlO.M animal undergoing GVHD from priited lyn^iiocytes : BIO .RIII anti-BlO .M. 143 30. Tbe lung of the BlO.M animal examined previously. 143 31. The lung of a normal irradiated mouse 14 days after irradiation 145 32. The lung of a BIO.RIII mouse reconstituted with prined BIO .RIII anti-BlO.M cells, day 14. 145 33. The spleen of the animal previously examined. 147 34. The liver of a sublethally irradiated BIO.RIII nouse reconstituted with primed BIO.RIII anti-BlO.M lyirphocytes 149 35. A hi^er magnification of the previous liver. 149 36. Mortality in sublethally irradiated (B10.T(6R) x B10.AQR)F, hosts induced by anti-I-A*^ long term cultured cell lines and clones 152 37. Inability of concentrated (ascites) anti-I-A*^ antibody (10.2.16) to prevent lethal GVHD in sublethally irradiated BIO. A mice reconstituted with B10.A(5R) splenocytes. 156 Xll

PAGE 13

FIGURES— Continued 38. Ths skin of a CBA/J newborn suppressed nouse day 72. 163 39. The liver of a CBA/J newborn G7H suppressed mouse day 72.-165 40. A higher magnification of the previous micrograph. 165 41. A single cell suspension prepared from a (BIO x B10.BR)F, spleen 25 days after reconstitution with BIO. BR cells and newborn splenocytes 17 4 42. Dose response kinetics of various suppressor factors en BIO.ER anti-C57BL/6 primary MLR. 182 43. Profile of sv:^pressor activity from newborn sapemate passed through a S^hacryl 300 column to inhibit a primary mixed lynphocyte reaction: BIO. BR anti-Bl0.D2. 185 44. Ouchterlony analysis of purified mouse AFP. 187 45. SDS-acrylamide analysis of the newborn factors. 189 46. Lethal GVHD in irradiated (C57BL/6 x BALB/c)F, mice reconstituted with BALB/c splenocytes. 195 47. Proposed model for GVH morbidity 234 Xlll

PAGE 14

Abstract of Dissertation Presented to the Graduate Council of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy THE IMMUNOBIOLOGY OF GRAFT VERSUS HOST DISEASE AND ITS ATTEMPTED PREVENTION USING NATURALLY OCCURRING SUPPRESSOR SUBSTANCES. By Martin Robert Jadus August 1983 Chairman: Dr. Ammon B. Peck Major Department: Pathology A model for the murine graft versus host (GVH) reaction has been developed in order to examine the potential value of naturally occurring suppressor cells and factors in preventing lethal disease. GVH reactivity was induced in sublethally irradiated adult mice by reconstituting the hosts with allogeneic splenocytes. Nonsuppressed host/donor combinations with genetic differences at H-2 died within 14 days. Combinations with I and K/D genetic differences showed longer survival times, while mismatches involving non-H-2 XIV

PAGE 15

generally had no mortality. T lymphocytes specifically reactive against host tissue antigens could be recovered from diseased organs such as spleen, lymph nodes and liver. Based on cell surface phenotype and in vitro reactivity, two distinct populations of cells were recovered: one proliferated against class II molecules, the other lysed cells expressing class I molecules. This situation proved different from reactivity profiles of cells activated in mixed lymphocyte reactions. The presence of cytotoxic T cells correlated with GVHD mortality. The presence of large numbers of CTLs was common in H-2 GVH mismatches and no doubt contributed to the rapid death. In contrast, only weak, but transient CTL activity was found in mismatches involving class II antigens; however, the recovered cells could be expanded in vitro with interleukin 2. The resulting cells exhibited cytotoxic activity against both donor and host. Attempts were made to prevent GVHD using cells from the spleens of newborn mice known to contain naturally occurring suppressor cell populations. Unexpectedly, only certain strains of newborn mice possessed the capacity to suppress lethal GVHD. The genetics of suppression by newborn spleen cells suggested two restrictions: first, the newborn spleen cells apparently must express the strongly stimulating Mis XV

PAGE 16

antigens, and second newborn spleen and adult donor cells must be histocompatible at a genetic region telomeric to I-A. No T cell reactivity could be detected up to 60 days in mice showing long term survival. During this time, these mice remained chimeras. After 60 days postengraftment the cell and reactive phenotype of these mice returned to that of the host XVI

PAGE 17

INTRODUCTION 1.1 The Graft Versus Host Reaction The graft versus host reaction (GVH) is a unique model to study immunoregulation in that the entire sequence of an immune response (initiation, differentiation, a multifaceted effector phase) and final control is represented. An extensive body of knowledge concerning the initial events of antigenic recognition, cell types involved, cell differentiation, and final expression of immunocompetence during the GVH reaction already exists (reviewed in references 1,2). However, there is comparatively little understanding of the complex mechanisms and cellular interactions regulating the course of the on going immune reaction ( s)

PAGE 18

The most common forms of GVH reactions are runt disease, secondary disease, parabiosis and F hybrid disease. Runt disease occurs when mature competent allogeneic cells are injected into an Immunol ncompe tent newborn. Growth of the newborn is inhibited and quite frequently the animal dies manifesting severe diarrhea, dermititis, hepatomegaly, and splenomegaly. ,aBw. Secondary disease is seen in those adult animals who have been immunocompromised by either drugs or irradiation and have been reconstituted with allogeneic stem cells with the following results: diarrhea, dermatitis, renal lesions (immune complex deposition), hemolytic anemia and hepato and splenomegaly. Parabiotic intoxication occurs when two allogeneic adult mice have been surgically treated so that they share a common circulation and their lymphocytes are free to attack the other animal. Hybrid disease happens when a F, hybrid has been injected with cells from the parental strain. The laws of transplantation (See Figure 1) state that when two histocompatible homozygous animals mate, the F progeny possess the histocompatibility antigens of both parents and are able to tolerate grafts from either parent, while the parent rejects grafts from the progeny because of the histoincompatible antigens. Thus the donor cells recognize

PAGE 19

Ul U) to M I ^ 3 cn 0) B) l-h ft W l-h i-( s ^d I l-h ft rt I a. CO rt 0) Ml (D O Hf5 PJ fD tt ^ fT ^ 0) s OJ 3 s m CD CO Oi ? f o S" (D N ft o" 'S (fl M •< to (D 0) cr D 1-1 l-h W HHO I &

PAGE 20

n c o c '~ 0) re ^ ^ ra 0) a c 0) o> CM U. m OJ E CNJ r>4 o 1 I • X o • L J L E o

PAGE 21

the F being foreign, but the F^ can not normally react to the donor cells because they see those cells as being self. The advantage of this system is that the recipient possesses an intact immune system. Currently GVH reactions are studied either systemically or locally. Both sets of reactions have been reviewed by Grebe and Streilein (2) and can be summarized as follows: A) Systemic Inhibition of syngeneic hematopoeitic colony forming units: Irradiated F, mice are given parental lymphoid cells along with syngeneic bone marrow cells. The amount of erythroid cell growth is assayed by the amount of Fe that is incorporated into the spleen. The less ^^Fe incorporated, the greater the immune response was towards the host. The Simonsen spleen index assay: This assay utilizies the fact that when allogeneic cells are injected into a recipient the spleen enlarges in response to the allogeneic challenge. The phagocytic index: This is an indirect test used in F^ animals 2 weeks after the parental cells are injected. Because of the increased lymphoid activity, colloidal carbon is cleared from the system in a much shorter time.

PAGE 22

The focal periportal infiltration method: This system deals with enumerating the number of foci seen in the F, liver after parental cells are injected. The splenic explant assay: Single cell suspensions of parental cells are placed over diced F, spleens. The culture is allowed to precede 5 days and then the culture is examined for increased physical masses. B) Local The epidermolytic reaction: sensitized lymph node cells are injected intracutaneously into F^ animals. The blood cells of the host are destroyed followed by nonspecific vascular destruction in the skin with noticeable epidermal necrosis. Intraocular or intrarenal GVH: allogeneic cells are injected either into the anterior chamber of the eye or under the renal capsule; within a short period of time immune reactions cause gross morphological changes in these organs. The popliteal lymph node assay: this assay is based on the same theory as the Simonsen spleen index except reactive allogeneic cells are injected into the animals footpad with subsequent measurements of the popliteal lymph node 1 to 2 weeks later.

PAGE 23

Often, the final result of GVH is the death of the host. At times, however, the host may survive GVH, suggesting the development of "tolerance". The mechanism(s) underlying tolerance, both host towards the graft as well as graft toward the host, remain only speculative, but may result from 1] blocking factors, such as antibody possessing the potential to bind to antigenic sites of the host cells (5,6) 2] suppressor cells, which inhibit the regulating T helper cells (7,8) or 3] destruction of the stimulating components of the host, presumably the lymphoid associated cells (9,10). The lymphoid cells of the host have been implicated in the histological and clinical manifestations of GVH in two ways. First, these cells no doubt provide the immunogenic stimuli for GVH through the trapping the donor cells within the lymphoid organs, thereby stimulating the donor cells into growth (11,12). Second, the host cells, while under GVH siege, may be nonspecif ically stimulated to grow (13,14). These proliferating host cells could either be target cells or they could become autoimmune to their self via an allogeneic effect mechanism (15,16) or via an inflammatory response. Attempts to alter GVH reactions have intensified with the development of human transplantation systems.

PAGE 24

J-vPharmacological agents have seen a major trend towards limiting GVH reactions. Cyclophosphamide and the cortical steroids have been tried clinically with less than expected results (17,18,19). A major problem with such drugs is that these agents nonspecif ically depress the immune system, thereby promoting other deleterious effects such as increased risk of infections and lymphomas. Cyclosporin A has recently become a center of a pharmacological approach to limit GVH, but it too may be nonspecif ically immunosuppressive (20,21,22). Another approach to this question has been to stimulate suppressor cells using the plant lectin, concanavalin A (Con A) (23,24,25,26) Certain doses of Con A are known to be mitogenic while other doses are known to be suppressive (25). Unfortunately, this drug also has toxic side effects and therefore could be hard to judge adequate dosages for any given individual. Treatment of donor cells with various antibodies and complement has been another effort which has been examined (27). Until quite recently, the only antisera which had been employed in humans was rabbit anti-lymphocyte antisera. The hope here was to eliminate immunocompetent T cells (while leaving virgin bone marrow cells intact) before they have had a chance to become stimulated. Unfortunately, this regimen has had only limited success which could be due to

PAGE 25

the poor specificity and titer of the antisera used. Perhaps the development of appropriate monoclonal antibodies might circumvent this problem (28,29). In any event, there is little disagreement on the need for better methods to achieve the final goal of tolerance. 1.2 The In Vitro Measurement of the GVH Reaction The murine major histocompatibility complex (H-2) consists of a number of immunologically important loci: K,D,S and I (Figure 2). The K and D regions code for molecules which direct lympholysis; the S region produces serum proteins such as the fourth component of complement, while the I region genes and associated antigens have been reported to be involved in a large number of immunological phenomena such as T-B cell cooperation (30), antigen presentation by macrophage (31), helper factor (32,33), suppressor factor (34,35) and blastogenesis in mixed lymphocyte reactions, MLR (36). Genetic disparity between donor and recipient which apparently controls initiation of GVH is determined primarily within the major histocompatibility complex (MHO of the species, in particular the I region (in the mouse) or its equivalent in other species. Representative data

PAGE 26

Figure 2. The major histcKxmpatibilty oomplex of the mouse. This schanatic drawing of the MHC of the mouse (H-2). The H-2 is composed of several loci, as denoted on tte figure. The MHC gene products is cotrposed of se\7eral types of molecules. Class I molecules such as: K, D and L nolecules are single 45,000 molecular weight proteins noncovalently associated with beta two microglobulin molecules. The class II nolecules: I-A and I-E consist of two noncovalently linked proteins of 35,000 and 28,000 daltons. Finally, the class III molecule encoded by the S region describes a serum cornponent C'4 v*iich is a oonporent of the oomplotent series of protein.

PAGE 27

w^ membrane U Chromosome 17 6 t Eg membronc Chromosome ? LZ membrane D Qa

PAGE 28

12 revealing the importance of the I region, taken from Klein and Park (37), is presented in Figure 3. Even though K and D region differences produce some splenic enlargement as evidenced by an increased splenic index of 1.5, this increase of 1.5 probably represents just the homing in of the injected spleen cells into the host spleen without any significant reaction. But when I region differences occur, the spleen doubles or triples in size due to the development of a significant reaction. The in vitro measurement of GVH is normally performed using the mixed lymphocyte reaction (MLR) (38,39,40,41). This test measures the proliferative phase of T cell activation following recognition of allogeneic antigens. Representative data comparing MLR and GVH are presented in Table 1 (37). In vitro studies using the MLR as a model for GVH have provided extensive insight into T cell recognition, differentiation, and cell interactions. Activation of T lymphocytes by MHC incompatibility results in increased DNA, RNA and protein synthesis, increased energy utilization and increased size (42,43,44). During this stage, the T cells proliferate and differentiate into either the proliferating helper T cell or the poised precursor of the cytotoxic T cell. The poised cytotoxic T cell further differentiates to

PAGE 29

Figure 3. Graft versus host reactions across various H-2 differences. This figure represents splenomegaly indices of various mice undergoing graft versus host reactions. Taken f ran Klein and Park ( 37 ) :!

PAGE 30

14 Meon Spleen INDEX ond S H-2 Region Difference of |q 15 2 2.5 3.0 SifOin Combinolion K Ir Ss-Slo Anim's ^ , , | , , ^ BIO.A >BIO Mi EBI MH PW '7/27 (=> AQR 5R ir~]l6/32 AR BIO BB 2R • AR CZ AR CZI •2R I BIO.A BIO.A No. of ID[ZI][IZ]<5/I9 I 20/22 II I|i20/i0 Difference in Region of H2 Origin | | 'No Difference in Given Region GVHR across //-2* differences. Numher of animals: control/e.xpenmental. No. H-2 Region Difference f | q Sffoin Combinolio n K Ir Ss-Slo D Anim's 1 , BIO.BR— > BI0.02 JH Hi MH H 12/17 2R BI0.D2 SH I 1 1 13/19 BIO.A — BI0.D2 M M I II 1 22/29 2R BIOHTG lMi <§ GU I 1 2^^^ BIO.AKf^-BIO.A I 1 1 I 19/20 2R BIO.A I II IF I — 18/ai Meon Spleen .Index ond SO 1.5 2.0 2.5 3.0 I I I ' I I I I I I i_ H- eiOBR WW pWi 16/25 AOR -BO BR Ml I |WMB i4/27 BI0.D2-* BIO.A Meon Spleen Index and S 1.5 2.0 2.5 3.0 —i — I — I I I I I I I 1—1 I I BIOHTG— 2R BIO AH AOR — BIO.A DC :c BIO.A— BlOBR r II I I BIOAKM'BIuBR I II 11 I I J 15/20 D 18/22 D 18/23 D 17/29 I 20/21 I 20/27 1-0-1 Difference in Rego t>f H-2'' Origin | | No Difference in Given Region (;VnR across //-2* (iilTcrcnccs. N'umhcr of animals: control/experimental. No. H-2 Reqon Difference of 1.0 SirOin Combinolion K Ir S^-Slo O Anim's L_i__i_ Meon Spleen Index ond SO BIO BIOG BIOA— 6R &0R — • BfO.G BiOA.AQR 15/21 18/25 20/27 CZI (ZZIIIZI 19/25 5R • Bio.G czii CD nm wm 14/23 1.5 I I I 2.0 2.5 -i—i—L J 3.0 -I ' I Difference in Region of H-2*' Origin | j "No Difference in Given Region GVIIU across //-J" differences. Number of animals: cimtrnl/c.xperiiuental.

PAGE 31

15 Table 1. Strenth of MLR and GVHR across various regions of the H-2 MLR GVHR H-2 region difference average ratio of stimulation range mean spleen index range K 2.0 .0.7-4.7 1.4 1.2-1.6 D 1.8 0.8-5.4 1.4 1.2-1.7 I-A 6.4 3.7-9.2 1.8 1.8 I-B+S 2.7 1.1-4.4 1.5 1.1-1.8 I+S 5.8 2.7-12.8 2.6 2.6 K+I 6.6 3.2-18.3 2.8 2.6-3.1 D+S 2.0 0.7-4.7 1.5 1.3-1.6 K+D 3.4 3.0-3.8 1.8 1.1-2.5 K+S+D 3.3 1.5-8.6 2.3 2.2-2.4 K+I+S+D 7.2 1.2-33.6 2.8 2.4-3.1 Taken from Klein (3)

PAGE 32

%s the mature killer cell after receiving the help signal from the helper T cell. Killer T cells were originally characterized as possessing the Lyt 2 and Lyt 3 antigens by negatively selecting T cells with anti-Lyt 1 antisera and complement (45,46). Recently, killer cells possessing the Lyt 1 phenotype have been described (47). The discrepancy in 2 killer T cell phenotypes may be due to the sensitivity in the antisera used. Originally anti-Lyt 1 antibody and complement killed the high density Lyt 1+ cells so that weakly positive Lyt 1+ cells still are present. Only through sensitive immunof lourescent techniques can one observe the low density Lyt 1+ cells (48). Helper T cells reacting against whole haplotype, I region, or non-MHC antigens have been shown to bear only Lyt 1 antigens (49,50). In contrast, T cells responding against K/D differences have been found to be Lyt 1,2,3+ cells (51). This as well may be an artifact with the Lyt 1,2,3+ cells being precursors to the T helper cells, Lyt 1+. The blast cells do not remain in this activated mode for long as they revert to small lymphocytes capable of exhibiting memory (52). These cells can be restimulated by the initial antigen in a secondary manner, with the peak response occurring on day 2 of culture as opposed to days 5 and 6 found in a primary reaction. The primed lymphocyte

PAGE 33

17 test (PLT) utilizies this concept of in vitro restimulation to specifically quantitate similar histocompatible antigens on third party cells (53,54). This procedure has only been developed in the last few years; however, the principles of these in vitro restimulations offer much potential in probing some basic questions of vital importance in immunobiology such as the activation of memory cells, the differences between primary and secondary stimulation, the specificity of the response towards antigens, and the number of different cell subpopulations responding towards one antigen. All these activities have been observed at the gross level, but the actions of single clonally expanded cells have only been recently approached. By selectively expanding blast cell cultures one may be able to examine the fine specific activities of individual cells which were not available previously. T cell clones have been established in various laboratories using a variety of different strains of mice as the source of responder cells. Essentially two types of functional clones have been found. One type of clone is cytotoxic towards the stimulating cell only when that specific antigen bearing cell and some exogenous helper factor(s) are present (55,56,57,58,59). Since these clones carry out this cytotoxic effector function, it is presumed that these cells originated from a cytotoxic T cell

PAGE 34

18 possessing Lyt 2,3 antigens. These cells may be maintained in culture for weeks solely by the presence of the exogenous factors even without the initial stimulating antigen. Interleukin 2 ( IL 2) is one of those exogenous factors; it is a 30,000 to 35,000 dalton protein derived from splenic lymphocyte cultures stimulated by T cell mitogens Con A or phytohemagglutinin (60,61,62,63,64), from other MLR reacting supernates (65,66,67) or from clones of T helper cells (58,59,68). The second type of clone undergoes proliferation when presented with the specific priming antigen, e.g., histoincompatible cells, or soluble antigens, but does not exhibit any cytotoxic potential. No exogenous factors are apparently needed by these clones. However, these cells can become addicted to IL 2 and can lose their specificity and ability to respond to antigen. A few of these clones have been shown to produce factors which activate the cytotoxic T cell clones in vitro (58,59), strongly suggesting that these clones are members of the helper T cell class. Helper T cell clones have also been shown to function in vivo by increasing antibody formation towards T cell dependent antigens, e.g., sheep red blood cells and horse red blood cells in nude mice (69). In addition to nonspecific helper factors such as IL 2, T cells have been claimed to produce antigen specific

PAGE 35

w factors which stimulate naive B cells into making specific antibodies towards a given antigen. Some of these factors bind to the antigen directly, possess la determinants, and have a molecular weight of 35,000 daltons. These factors when isolated from T cells can direct the B cells into antibody secretion (70,71,72,73). Whether a given T cell secretes both IL 2 and antigen specific helper factor is not known The majority of cytotoxic clones which have been established have activity against specific membrane moieties, like H-2K or H-Y antigens (74,75,76). In contrast, the majority of proliferative helper T cell clones have activity against non-MHC antigens (77,78), although a few have been selected with activity against H-2 I region gene products (la) (79,80). One of the problems with long term lymphocyte cultures and clones has been the frequent and regular appearance of a crisis phase in the growth pattern of these cells (68). After 3 to 6 months in culture the majority of the responding cells die; the remaining cells replicate at a slower pace than they did before crisis set in. This crisis period lasts for about 3 weeks, after which time the cells may once again grow. Thus, long term cultured cells can be used for analysis only during the limited periods between crises.

PAGE 36

20 1.3 Controlling the Activities of T Cells In Vitro Alpha fetoprotein (AFP), a normal component of fetal and newborn sera, has been shown in both human and murine systems to exert selective suppressive effects on various functions of T lymphocytes, including T cell dependent antibody synthesis, T cell mitogenic responsiveness, and T cell mediated allogeneic reactivity (81,82,83,84,85,86). In addition, recent reports have revealed that under certain circumstances AFP may exert a supportive influence on in vitJ^o cell growth with one manifestation being the in vitro induction of suppressor T cells (83). Analysis of the impact exerted by AFP on the recognition and subsequent proliferation of T lymphocytes reacting in MLR against defined histocompatibility alloantigens has revealed a highly selective activity in the suppression of lymphocyte responses (81,82). In general, AFP inhibits Lyt 1+ T blast cells reacting against I region structures, including reactions against Mis locus products, but fails to inhibit Lyt 2+ cells stimulated by K/D alloantigens (81). Thus, it seems clear that AFP exerts its suppressive activity in MLR via selective interference with I region triggering systems. However, AFP also suppresses the effector phase of the T cell mediated cytotoxic reaction, thus suggesting a broader spectrum of regulatory

PAGE 37

21 activity. For example, AFP, when present during the primary activation phase of T cell responses, not only suppresses the subsequent in vitro generation of effective cytotoxic T cells in strain combinations with I plus K/D region differences, but also in strain combinations possessing only K/D region differences where the proliferative phase was unaffected. If AFP interferes only with I region triggering, then it would have been expected that at least in reactions directed against isolated MHC SD region associated gene products not only the proliferative but also the cytotoxic phase would have remained refractive to the suppressive activity of AFP. More recently, studies by Peck et__al. (87) have shown that AFP acts on the stimulating cell population known to initiate T cell reactivity. Furthermore, the T cell subpopulation which is refractive to AFP could be shown to exert suppression of normal primary responses. This fact suggests that AFP may be a physiological substance which could be used to control specifically the immune response in GVH reactions. 1 4 Suppressor Cells 7 A subpopulation of cells has been described which /' resides in the spleens of the newborn mouse which possesses \ a short lived antigen nonspecific suppressor activity for

PAGE 38

22 / immune reactions (88,89,90). This suppressor activity is not associated with spleen T cell populations as it is absent from purified spleen T lymphocytes, resistant to treatment with anti-I-j and anti-la antisera, present in the spleens of T cell deficient nude mice. In addition, suppressor activity is not due to macrophages since the effector cells I fail to adhere to either plastic or Helix pomatia lectin \ N coated plates. Natural killer cells can also be excluded since the suppressor activity fails to pass through Ig anti/ Ig coated columns (Peck, unpublished results). In addition, the cells from those animals are capable of producing a soluble factor which is capable of inhibiting adult cell responses both MLR and CML. ^Another set of suppressor cells has been postulated (2) to be responsible for the inability to transfer GVH from a host animal undergoing GVH reaction to a second normal animal even when both animals are genetically identical. Perhaps that time required for secondary reactivity to establish itself in the second host may permit an antigen specific suppressor cell to develop. Both of these activities need to be further investigated as physiological entities to suppress immune responsiveness.

PAGE 39

23 1.5 The Rationale for These Experiments GVH reactions have been studied in mice in a wide variety of ways, ranging from local footpad swelling assays to splenomegaly studies in newborn hosts. Unfortunately, the majority of these studies are not homologous to the situation seen in human GVH following bone marrow transplantation. Studies dealing with the homologous situations in adult mice undergoing GVH can be classified into two main categories: first, those which have dealt with the histopathological lesions, and second, those studying the ability of cytotoxic T cells to develop using an entire H-2 mismatch. The purpose of this research was to develop a model to study GVH reactions resulting from specific H-2 and non-H-2 incompatibilities between donor and host in order to study the genetic control of T lymphocyte reactivity in GVH disease, then determine the feasibility to control this reactivity with physiological pregnancy associated substances.

PAGE 40

MATERIALS AND METHODS 2 .1 Animals Inbred lines of mice used in these studies and maintained in the Department of Pathology, University of Florida, include A/J, AKR/J, A.TFR5, A.TL, BALB/cJ, BALB/C BIO. A, BIO.AQR, B10.A(2R), BlO.AOR), B10.A(4R), B10.A(5R), BIO. BR, B10.BUA16, B10.CHA2, B10.D2, BIO.F, BIO.HTT, BIO.M, B10.M(17R), BIO.MBR, BIO. PL, BIO.RIII, B10.S(7R), BIO.SOR), BIO.SM, B10.T(6R), BIO.Q, BIO.TL, CBA/CaH, CBA/J, C3H/HeJ, C57BL/6J (B6), C57BL/6^"'-'-, C57BL/10 (BIO), C57BR, C58, Dl.C, DBA/lJ, DBA/2J, PL/J, NZB, RF/J, SEA/J, SEC/J, SJL/J, SM/J and SWR/J. Breeding pairs of BIO.GD, BIO.RIII, B10.S(7R) and B10.S(9R) were originally provided by Dr. Duncan, Department of Cell Biology, University of Texas, Dallas, Texas, while BIO.AQR and BIO.TL were obtained from Dr. Shreffler, Department of Genetics, Washington University, St. Louis, 24

PAGE 41

25 Missouri. BIO males were bred to BIO. BR females, A.TL males bred to BIO.MBR females, and B10.T{6R) males bred to BlO.AQR females provided the F, hybrids, (BIO x B10.BR)F (A.TL X B10.MBR)F^, and (B10.T(6R) x B10.AQR)F Both male and female mice ranging in age from 6 to 24 weeks were used; however, mice were sex matched when used for various experiments. 2.2 Antisera Monoclonal anti-Lyt, anti-Thy-1 and anti-I-A antibodies were obtained from cell lines 53-7.313 (anti-Lyt 1), 53-6.72 (anti-Lyt 2), HO-13-4 (anti-Thy-1 2 ) and 10-3.6.2 (anti-I-A ) generously provided by Dr. Ledbettor and Dr. Herzenberg via the Salk Institute Cell Distribution Center, San Diego, California, while anti-K (B 312) and k anti-I-E (14-4-4) were provided by Dr. D. Sachs, NIH. Each cell line was grown at high concentration in RPMI 1640 supplemented with fetal bovine serum to 10%. Supernates were used undiluted. The antibodies 10.2.16, B 312 and 14-4-4 used in this research were generated in (B6 x BALB/c)F, mice using the ascites approach. These antibodies were demonstrated to high cytotoxic titers and were generously provided by Dr. E. Wakeland.

PAGE 42

26 Arsenilic acid conjugated anti-Lyt 1 and anti-Lyt 2 antibodies and fluorescenated rabbit anti-arsenilate antibody were obtained from Becton-Dickinson, Oxnard, California. Fluorescenated rabbit anti-mouse Ig antibody was obtained from Cappel Laboratories, Cochransville, Pennsylvania. 2.3 Complement Dependent Antibody Cytotoxicity A two step complement dependent cytotoxicity assay was used to treat cell populations with various antisera. Spleen cells at 5.0 x 10 cells/ml were incubated with antiserum for 45 min. at room temperature. The cells were then washed, resuspended in rabbit complement (Accurate Scientific, Hicksville, New York) and reincubated at 37 C for 45 min. Cell viability was assessed by trypan blue dye exclusion. 2.4 Fluorescent Microscopic Determination of Cells Immunofluorescent staining of lymphocytes was performed by reacting 5.0 x 10 lymphocytes with either arsenilate conjugated anti-Lyt 1 or anti-Lyt 2 antibodies for 45 min. at 4 C. The cells were washed 3 times with phosphate buffered saline (PBS) followed by an incubation with fluoresceinated rabbit anti-arsenilate antibody for 45 min

PAGE 43

27 at 4 C. The cells were examined for fluorescent staining through a phase contrast microscope equipped with a Zeiss Ploem UV illuminator. Cell surface immunoglobulin was detected in a similiar manner using a one step incubation with fluorescein conjugated rabbit antimouse immunoglobulin. 2 .5 Cell Preparations Whole spleen leukocyte populations were prepared as 'described by Peck and Bach (91). In brief, spleens freshly removed from mice were dispersed by pressing the spleen / through a wire mesh screen into PBS. Following one wash, the I red blood cells were lysed in a 10 minute 0.84% ammonium chloride treatment. The resulting leukocytes were washed once and resuspended in PBS to appropriate cell concentrations 2.6 Primary Mixed Leukocyte Reaction / :' / 7 Primary MLRs were carried out according to the protocol of Peck and Bach (91). Throughout the study, 60 x 10^ splenic leukocytes were cultured together with 100 x 10^ (2000R) stimulating whole spleen cells in 30 ml EHAA media supplemented with normal mouse serum to 0.5% (92). Cell cultures were harvested between days 7 to 10 of incubation

PAGE 44

28 [yj and examined for reactivity in secondary MLR (PLT) and cell mediated lympholysis assays (CML) 2.7 Induction of Graft Versus Host Reactivity Forty million donor (responding) splenic leukocytes were injected intravenously via the tail vein into sublethally irradiated (650R) recipient (host) adult mice. Animals were fed on lab chow and given acidified water to drink. No major problems developed from bacterial infections. At various time points (indicated in the text and footnotes) the recipient mice were sacrificed. Their livers, spleens, kidneys, intestines, lungs and skin were removed, fixed in formalin and embedded in paraffin for routine hemaoxylin and eosin stained histopathological study. Cells present in the spleens, lymph nodes and livers were prepared and examined for functional activities. 2.8 Primed Lymphocyte Typing Tests Antigen activated cells were separated from small and dead cells on Ficoll-Paque (Pharmacia Fine Chemicals, Piscataway, New Jersey) (d=1.077) density gradients. PLT assays were performed in round bottomed 96 well Lindbro '^
PAGE 45

29 microtiter plates (Flow Laboratories, McLean, Virginia) as described by Peck and Wigzell (93). Cultures consisted of either 0.04 x 10 responding spleen, lymph node or liver cells cultured with 0.5 x 10^ irradiated stimulating spleen cells. At appropriate times of the secondary culture, as indicated in the Tables, 1.0 |aCi tritiated thymidine 3 ( H-TdR) (Amersham, Boston, Massachusetts) in a volume of 0.02 ml was added to each well for 12 hrs. Cells were then aspirated through Whatman glass fiber filters with a multiple sample harvester (Otto Hiller Inc., Madison, 3 Wisconsin) and total H-TdR incorporation was determined by liquid scintillation procedures. Data are expressed in counts per minute of the average of either duplicate or triplicate cultures. Standard deviations of the average are included. 2.9 Cell Mediated Lympholysis Assays CML were performed according to the procedure detailed elsewhere (94). The effector cells were generated in primary MLR or GVH. Alloantigen activated cells were collected from mixed lymphocyte reactions or diseased organs, centrifuged and washed twice in medium containing newborn calf serum (Biocell Laboratories, Carson, California).

PAGE 46

30 Concanavalin A (Pharmacia Fine Chemicals) stimulated spleen cells were used as the target cells. Approximately 36 hrs before the CML assay, appropriate target cell cultures containing 12 x 10 spleen cells in 5.0 ml of EHAA with 10% newborn calf serum were established in culture dishes. Target cells were incubated 2 hrs with 400-500 |aCi Na^ CrO^. The labeled cells were washed 3 times in fresh medium. Cell destruction was performed in V bottomed microtiter plates (Lindbro) using various effector cell numbers plus 4 1.0 X 10 labeled target cells. Cell destruction proceeded 6 hrs at 37 C after which time the plates were centrifuged, the supernate collected and the quantity of released of Cr determined. Percent cytotoxicity is expressed as Cr released (experiment)— Cr released (spont.) xlOO Cr released (maximum) Cr released (maximum) Spontaneous release ranged between 15 to 20% of the maximum release. 2.10 Preparation of Purified Interleukin 2 Interleukin 2 was produced by culturing EL-4 G-12 cells, an azogaunine resistant EL-4 (T lymphoma cell line) with 8 ng/ml of concanavalin A for 18 hrs as described by Shimizu et_al. (95). The interleukin 2 was collected and

PAGE 47

31 precipitated by ammonium sulfate and subsequently dialyzed in PBS. The supernate was concentrated and passed through a G-200 column. The fractions containing the IL 2 were collected and concentrated. These samples were then sterile filtered. The activity of the preparation was tested by using an IL 2 dependent cell line by Dr. Shimizu. • 2.11 Preparation of Newborn Spleen Ce lls C. Newborn mice 1 to 3 days old were sacrificed; their y spleens were removed using sterile techniques. The spleens were passed through a fine wire mesh screen to get a single / cell suspension. The cells were washed once with PBS and then caltured -la ^HAA media at a concentration of 10 x 10^ cells/ml'. Normally 2 x 10^ viable cells are obtained per (spleen. To generate suppressor cells used for the prevention of GVH disease, the newborn cells were cultured 1 day prior to the addition of the donor cells. This mixture of cells was allowed to incubate an additional day. These cells were harvested, counted and injected into sublethally irradiated host mice.

PAGE 48

32 2.12 Preparation of Alpha Fetoprotein Pregnant mice between (10 to 14 days) were cervically dislocated and surgically opened. The amniotic sac was punctured using a needle; the amniotic fluid was then aspirated into a collection flask. The amniotic fluid was passed through an affinity column of rabbit anti-mouse AFP antibody generously provided and established by Dr. A. Kimura. The AFP was eluted off the column using glycine-HCl buffer pH 3.5. The AFP was subsequently dialyzed in PBS three times in 1000 x volume. The AFP was then tested in a primary MLR to insure that the collected substance had suppressor activity in it.

PAGE 49

RESULTS 3-1 Inability of Allogeneic Cells to Induce GVHD in Normal Adult F -. Mice Adult male and female (BIO x B10.BR)F mice were injected either intravenously or intraperitoneally with forty million BIO. BR splenocytes. Each set of mice was observed for a period of 1 month, during which time no signs of ill health were apparent. This experiment was repeated on a new set of mice, except this time the experiment was terminated on day 5 in order to examine the spleens of these animals. The mice which were injected intraperitoneally showed no signs of illness; the spleens of these animals looked normal and possessed a normal number of leukocytes (60 X 10 cells/spleen). In contrast, the mice which were injected intravenously exhibited splenomegaly and had twice the number of leukocytes, 120 x 10^ cells/spleen. Despite this enlarged spleen, the animals looked and acted like normal mice. 33

PAGE 50

34 The spleens of these two sets of animals were prepared into single cell suspensions and forty thousand cells were dispensed into each well of a microtiter tray. The cells were then tested with a panel consisting of irradiated spleen cells from five different mouse strains as is done in a typical primed lymphocyte typing test. The cells from mice which were injected i.p. did not significantly respond towards any of the stimulating cells (Table 2, column 1). In contrast, those cells obtained from the i.v. injected mice looked larger and appeared activated. These cells did respond towards the panel of stimulating cells (Table 2, \ column 2); however, it was without any specific pattern. 3 2 Generation of GVHD in Adult F Mice Immunosuppressed Through Irradiation A. Study of the survival rates of mice lethally or sublethally irradiated To obtain conditions in host animals permitting development of GVHD, it was necessary to immunosuppress the host through irradiation. A dose of irradiation was desired so that the animal would not die of infection due to the associated leukopenia, yet be compromised enough so that the donor cells would be opposed with the least possible resistance.

PAGE 51

Table 2. The effect of the route of injection in order to generate primed lyit^jhocytes in the BIO. BR anti-(B10 x B10.BR)F reaction. 35 H-T'dR Incorporation CPM+SD (a) H-2 Genetics Stimulator Mice injects in traper i toneal 5d by: (b) Strain K A J E D intravenous none 173+ 66 8570+ 107 BIO. BR k k k k k 622+298 7551+1402 BIO b b b b b 525+ 4 10875+ 341 (BlOxBlO.BR) b b b b b ^1 k k k k k 679+232 8046+ 6797 BIO .T( 6R) q q q q d 711+711 8651+ 75 DBA/2 d d d d d 813+ 95 9965+ 835 a) PLTs v^re harvested at 48 hrs following a 12 hr pulse with 1 pCi of H-TdR. b) Primed cells were obtained on day 5 from sublethally irradiated (BIO x B10.BR)F, mice v^iich were reconstituted with BIO. BR splenocytes by either an intraperitoneal or intravenous route.

PAGE 52

36 Figure 4 shows the survival pattern of groups of 4 mice which were irradiated with either 475, 650, 875 or 1100 rads. A dose of 475 or 650 rads produced no observable detrimental effect up to 75 days post irradiation when the experiment was terminated. Higher doses of radiation used (875 or 1100) caused death in these animals by day 14. Thus, lethal irradiation induces quick mortality while sublethal irradiation has no obvious detrimental effect on the mice. B. Sublethally irradiated hosts provide an envi ronment for GVHD ~ To test the effect of irradiation on the host in order to generate primed lymphocytes, (BIO x B10.BR)F host animals were divided into two groups with one group receiving 650R while the other group did not receive any irradiation. The host animals were injected i.v. with 40 x 10 BIO. BR splenocytes. Five days later these animals were sacrificed, their spleens were removed, and the cells tested for alloreactivity in MLR. The results of one experiment are reported in Table 3. Cells from animals which were not irradiated produced cells which did not respond specifically towards any particular strain of mouse. However, cells from those mice which were sublethally irradiated showed patterns of reactivity suggesting specific reactivity against the

PAGE 53

n c o *-• 4rf "y nj O O 0) •H 0) "O tn (u n Vj 3 Q S V^ O 'O CJijj ly C O ly (1) O H >i 4J 0) nj O 05 u-i ^3 d; Q) <4-i (0 •^ r-4 13 O > 'O ^§

PAGE 54

38 (A D w m h(A o O o in nvAiAans iN3Da3d

PAGE 55

3d Table 3. The effect of radiation cmi the host in order to generate specific primed lynphocytes in the BIO. BR anti-(B10 x B10.BR)F. reaction. ^ Stimulator Strain H-2 Genetics K A J E D none _ _ BIO. BR k k k k k BIO b b b b b (BlOxBlO.BR) b b b b b B10.T(6R) DBA/2 k k k k k q q q q d d d d d d H-TdR Incorporation CPM+SD (a) Ccxidition of Host: (b) Not Irradiated Irradiated 8570+ 107 7551+1402 10875+ 341 8046+ 697 8651+ 725 9965+ 835 1947+ 133 1429+ 252 42062+1809 44871+ 402 6146+ 292 10168+2227 a) PLTs vgere harvested at 48 hrs following a 12 hr pulse with 1 [iCi of H-TdR. b) Primed cells were obtained on day 5 frccn either sublethally irradiated or non irradiated (BIO x B10.BR)Fhosts which were reconstituted with BIO. BR splenocytes injected i.v.

PAGE 56

40 antigens which were found on the host. Thus, by immunocompromising the host it is possible to generate a GVHR C. Histology of sublethally irradiated (650 rads) mice Various tissues from animals which had been previously irradiated 10 to 15 days earlier were removed, prepared for thin sectioning and then examined for abnormalities. Liver, kidney and small intestines appeared normal. Lymphoid tissue such as thymus and spleen were atrophic. The spleen contained 0.05 x 10 cells and the histology of the spleen revealed necrosis, but did show signs of regeneration with the presence of megakaryocytes D, General pathology of GVHD in MHC disparate strains Sublethally irradiated animals reconstituted with 40 x 10 allogeneic or semi-allogeneic splenocytes developed acute GVHD. As early as five days postgraf ting, mice started to deteriorate physiologically: The mice became lethargic, developed hunched postures, presented a wasting appearance and developed diarrhea. Often immediately before death the mice felt hypothermic and were shivering. In semi-allogeneic combinations such as (BIO x B10.BR)F, mice reconstituted

PAGE 57

41 with BIO. BR cells, mice began to die by day 7 and by day 14 to 15 all the mice had expired. In allogeneic combinations such as the 36 anti-BALB/c reaction an acclerated course of the disease was seen, and by days 5 to 6 all the mice had died. A comparison of the survival rates due to different histocompatibility loci is presented on Figure 5. An entire H-2 disparate GVH reaction [B6 anti-BALB/c] resulted in a very short lifespan. An I region mismatch [BlO.AQR anti(B10.T(6R) X B10.AQR)F ] produced a slightly delayed mortality, while a K/D disparate GVHD [BlO.MBR anti-(A.TL x B10.MBR)F^] resulted in about half the mice surviving the first 20 days. E. The effect of host's age upon generation of GVHD Another study was undertaken to determine whether age of the host had any influence on the rate of GVH mortality. In one experiment an entire H-2 combination was used: BIO. BR anti-BlO.WB. The age of the hosts ranged from 22 days (mice are normally weaned on day 21) to 70 days old (young adult). As can be seen in Figure 6, the majority of the mice died between days 4 to 8 and by day 10 all the mice had succumbed to GVHD. Thus, lethal GVHD developed similarly in mice

PAGE 58

X ^3 -P 4J • >i W (0 O _n3 O -H >-l TJ w 'o m • >-i vj is •H ^1 o +J B '^ ^ c rH >iCii O >i.H < T3 -H J2 O S-< tH 4-) iH O •U 3 Vp M M CQ
PAGE 59

43 or o u. < or O m GO S X o u J3 10 CD X o m 1d _l 1. \o m 1" la rr q: •a c -1 o m a> a> m < ^ E> X = o 9} o V -y* "o o> n T3 5 = tz 5 H \ ^
PAGE 60

B ^ '^ • O O o •m iH fO 'U -H (!) C CM C •H CM a' • n\t3 p w o a; P 21'-' -J U (0 (o ^

PAGE 61

45 -1 pj (O = o Q a o o 2 _] _i -I -I -J o o o o o •5 to (o (n o >->->->-> X < < < < < a a a a o N O CO — o cvj N lo m K g 4 • O IZ < z UJ Iv> o Q. w >< Q nvAiAans iN30d3d

PAGE 62

46 y _^. ranging in age from 22 to 70 days old. Experiments performed in this study thus used mice which were this age. P* Histopathological examination of whole H-2 disparat e GVHD revealed marked effects The pathology described here is similar to that previously reported by Rappaport et__al. (96). The livers of these GVH affected animals often changed from a normal red color to a pale white coloration. This condition seemed to frequently occur before any wasting syndrome presented itself. These livers showed signs of perivascular cuffing, dilation of the veins, massive necrosis with no signs of regeneration. There was marked evidence of leukocytic infiltration in the parenchyma as well as along the central veins (Figures 7 to 10). The yield of leukocytes from the GVH liver varied in different experiments but usually 20 x 3 4 10 to 70 X 10 cells/liver could be recovered. In contrast, normal livers failed to yield any significant amount of leukocytes. Similarly, examination of the small intestine revealed drastic changes: the villi were dilated with columnar metaplasia. Exfoliation of the villi was also noted to be higher than normal. Leukocytes were devoid in the villi (Figures 11 and 12). Destruction and necrosis was obvious, and leukocytic and plasma cell invasion of the basement

PAGE 63

Figure 7. Mouse liver from either a nornal aninal or fron a sublethally irradiated animal on day 10. Both livers exhibit no drastic dianges morphologically. The liver is uniformly packed vath hepatocytes. Central veins are seen at lower left and center. A bile duct is seen in the qpper right. Figure 8. Normal mouse liver, a higher magnification. Tte previous section was examined londer higher magnification. The artery is situated in the center. Notice tte uniformity of the cells. Staining the cells with hematoxylin and eosin reveals the hepatocytes have an eosinophilic cytoplasm with well defined cell membranes.

PAGE 64

*^ -t v nr41 •; v\ V .• 0' .. iV^^*\

PAGE 65

Figure 9. The liver of a mouse undergoing acute GVHD. This liver was obtained frcxn a (BIO x B10.BR)E^ mouse undergoing acute GVHD induced by 40 x 10 HBIO.BR splenocytes at day 10 of the reaction. Perivascular cuffing exists along the central vein. Leukocytes can be seen along the vein as well as infiltrating the parenchyna. Tte normal architecture of the liver appears to be disturbed with nunerous vacoules found in the parenchyma. Figure 10. A higher magnification the previous liver. Leukocytes have infiltrated along the bile duct and have invaded the parenchyma. Notice tlie loss of normal cellular distribution. The hepatocytes show ooagulative necrosis, tte cytoplasm of the oells has been disrupted with only nuclear remnants left.

PAGE 66

St,

PAGE 67

Figure 11 Tte intestines of a normal itouse. Tba intestines of a normal mouse are similar to those found in a sublethally irradiated mouse at day 10, although there is a slight (tecrease of (10 to 20%) leukocytes in the villi. The intestine here shows the villi are intact and have leukocytes along the lacteal s. Figure 12. The intestines of a mouse undergoing acute GVHD. A sublethally irradiated (BIO x B10.BR)F, mouse was injected with 40 x 10 BIO. BR splenocytes and vas examined on day 10 of the reaction. Epithelial cells are still along the periphery of each villus. Necrosis is observed in the epithelium. Tte lacteals are remarkedly devoid of leukocytes (90 to 95%).

PAGE 68

cJ2 '•• ' • • Mf: 'tv *%* \ p^i ij I • ^. A; .^tlf,/: f^ £^^ H* \:Kh ^. f^ f • X i I' "^ 5o '-^ ^ '^ r^ ~' f^ i# <^\Jl ^:-' k/v^

PAGE 69

53 membranes was pronounced. Several attempts were made to extract leukocytes from these tissues, unfortunately, no cells were recovered. Histopathological examination of the whole H-2 GVH spleens on days 6, 10, and 15 days postgrafting revealed marked atrophy with disruption of the normal white pulp architecture, similiar to results found in the sublethally irradiated mice (Figures 13 to 17). However, more viable cells (5 to 15 X 10 cells/spleen) could be recovered from these types of spleens. G. Pathology of I or K/D region disparate GVHD In I region GVHD the spleen and liver appeared to have leukocytic infiltrates which progressed with time. The predominate cell type found in the spleen was the polymorphonuclear leukocyte (PMN)(70 to 90%). The intestines developed abnormalities in the second week after engraftment. The lesions which were seen in these animals were never as severe as those seen in whole H-2 disparate GVHDs. In addition, the pathology of one animal in a given series of GVHD was at times dissimilar to that seen in another animal indicating advanced stages of disease occurred in some animals but not in others. Mice died between 8 to 17 days after engraftment (Figure 5).

PAGE 70

Figure 13. The spleen fron a normal mouse. Tte germinal centers are slightly hypercellular but the normal architecture is intact. Figure 14. The normal spleen, a higher magnification. This section here displays two germinal centers.

PAGE 71

55

PAGE 72

Figure 15. The spleen of an aniital undergoing acute GVHD. A sublethally irradiated (BIQ x B10.BR)F, mouse was reconstituted with 40 x 10 BIO splenocytes. The aninial was sacrificed on day 10 cf the reaction. The overall architecture of the spleen has been disrupted and appears to be identical to that produced by a sublethal dose of irradiation. The red pulp appears to be repopulated by cells. Figure 16. A higher magnification from the previous tissue. Areas are fibrosed with collagen deposition and have numerous cells in the area. A great majority of the cells appear dead and this is confirmed v^en a single cell suspension is examined using trypan blue dye.

PAGE 73

57 ^^i^'^tfe*?=: ^7* i^S?^t2!£^*^S* *-5 ?3pJ

PAGE 74

Figure 17. Further magnification of the previous spleen. The cellular infiltrate of this area appears to be nononuclear in origin. Massive amounts of necrotic cells appear to be observed in the center of the field. Figure 18. A cytocentrifuge preparation of cells obtained from a itouse spleen undergoing acute GVHD. When the GVH spleen is passed through a wire screen and the cells are passed over fiooll gradients, to ronove the dead cells, the reitaining viable cells appear to show a variety of different cell types, this figure shows that about 50% of the recovered cells are PMNs. Several lynphoblasts are found along with nurrerous snnall lynphocytes.

PAGE 75

59

PAGE 76

60 In K/D region disparate GVHD, pathological conditions similar to those found in I region GVHD were seen; only the symptoms of the disease seemed to be delayed or absent. Death usually occurred from days 10 to 20 and some mice survived longer than four months (Figure 5). Infiltrates were found in the spleen and liver; again the predominant cell type recovered was the PMN (70 to 95%). Intestines were frequently normal; very few villi were dilated and depleted of leukocytes. However, about 60% of the animals did die in an emaciated condition. H. Cellular composition of the host organs und ergoing GVHD Cytocentrifuge preparations of single cell suspensions of anti-H-2 GVH spleens revealed 40 to 50% of the recovered viable cells were PMNs, while 50 to 60% were lymphocyte/monocytes (Figure 18). Leukocyte preparations extracted from the livers revealed a similiar composition. Similar preparations of anti-I or anti-K/D region GVH spleens contained 70 to 90% PMNs. Culturing the leukocyte preparations overnight in tissue culture medium resulted in the majority of the PMNs dying, thereby facilitating isolation of the lymphocyte/monocyte population on ficoll-isopaque density gradients (d=1.077). Of the remaining viable cells, 30 to 35% were immunoglobulin

PAGE 77

61 positive (determined by cell surface immunofluorescence); 55% stained for the Lyt 1 marker, while 20% stained for the Lyt 2 marker. Monoclonal anti-Thy 1.2 antibody plus complement killed 60% of the cells providing evidence that the majority of the cells were T cells bearing the Lyt 1 marker. IRecovery of viable lymphocytes from various histoincompatibility differences The number of viable lymphocytes out of the spleen varied depending upon the genetic disparity. Table 4 summarizes these findings. The spleens of the hosts undergoing GVHD were removed on day 5 of the GVH reaction and were incubated overnight; the remaining viable lymphocytes were then enumerated the next day. In general, those combinations with the maximum genetic mismatch (entire H-2 mismatched reactions: combinations 1 to 3 ) yielded the most lymphocytes/spleen: 2.5 to 5.1 x 10^. Whereas in those combinations which differed in class I molecules produced between 2.2 to 4.6 x 10^ lymphocytes/spleen (combination 6 and 7). Class II disparate reactions yielded varying numbers of lymphocytes. Those combinations which were directed against the I-A molecule yielded the most cells/spleen (combination 8 and 9) 2.8 to 4.1 x 10^, while (BIO.GD x

PAGE 78

62 (0 8 Q o iH ir • • • in in oj 00 iH tH XXX o in m • • • o VD r^ ^ in CM XiTi M CN CM (N I I t £ S S fe g s • o • •-\ o CQ •-i 1 ^ s § m o ^ •T CQ fa rH ^ ^^ o S O" S 1 o •H -H • (0 O o fH i-H <^ CQ CQ OQ ^ o X in o X <: I Ui o o iH X o .H CM CO fa ^ 8 o ^ OQ 33 ^ -H BJ o '-4J CTi i-H '*. g55S S o CQ rH w ocj ^ in in o X o rH X ro Q fa I •H 4J CQ vo in o iH X CN (N VO O X CM CM u. I OS rH o rH CQ vD vD in in o o o o in o X X X X X rH 00 00 CM CO • • • • • "afS T rvo O iH O VD VO VD VD O O O VD O XXX X o r^ 00 n • • • • iH O •^ TT VO o rH X in CM f^fiTfi M H H M H r-O fa rH ^-% CQ • jj s ^ S a ^-^ • • OS o o CTi rH iH o O rH ^ CQ OQ ^ 1^ -H 4J fa CM CU o rH I Qi H VO O oJ — o 8 §PQ o • r-4 O OQ -H — CQ 00 ^ O iH Cn rH rH < o rH CO <1> V-l O C T3 XI 5 0) O rH "O 0) 'Q 8 ^ OJ o 'is 1 4J 0) W rH ,H •§ s (0 Sj (U 4-1 w 4J E-l c C N 0-H •H rH CO -H C 4J IS 4J B -H 01 to ja CO (0 BO) 0) H 4J O 5-1 §&^^ O -H 4J C W 8r^-

PAGE 79

63 B10.MBR)F^ anti-BlO (combination 10) gave the least, 4.8 x 5 10 cells/spleen. Thus, it appears that I region GVH reactions are the most variable in terms of obtaining viable lymphocytes from the fifth day of the GVH reaction. J. Functional activities of leukocytes obtained from GVH animals The lymphocytes recovered from these GVH affected tissues were then tested in functional tests to determine what antigen(s) the recovered lymphocytes are capable of reacting. Two functional tests have been used: 1) the mixed lymphocyte reaction/primed lymphocyte typing test which measures the ability of reactive cells to proliferate against foreign histocompatibility antigens, 2) the cell mediated lympholysis test which assays the ability of primed cells to lyse Cr le appropriate antigens. cells to lyse Cr labeled target cells possessing the i. Mixed lymphocyte reaction/primed lymphocyte test. BIO. BR anti-(B10 x B10.BR)F^ reaction The spleen cells recovered from the (BIO x B10.BR)F, mice undergoing GVHD induced by BIO. BR cells were found capable of proliferating in a mixed lymphocyte reaction. The kinetic responses of these lymphocytes are presented in Figure 19. The activated cells reacted to BIO (H-2'^)

PAGE 80

to -rH u H OJ m oj • Q u u-i o • o O .H o iw ::^ -V (u x: g m 4J y Eh C -H o I as K cm CJ 43 J o 3 CQ O jJ r>•-! •H -H in u tn 4J 4J U li. 0) a) w +J C C v,H j^ 8 S x: • 4J c Q) >-i o -H nH 4J Q) CQ a o Tj w w -H .. m (1) E >-i to fl) (0 HC •H 4-) '^ ••-( O • JS W S Q +J a c CQ >y (1) X T) O >^ fc rH m •H '^ n 4-1 r-l +3 o X B -S "^^ CliO M-l a W (U CQ T3 fl) rH in iU >(!) jC m Eh I C Eh Q • 4J (C • J-l i'-j >i M m tn m d m M r-H > g 'tJ &IO r-t -H •rH 1-t n) C 4J C fa CQ O O CO O 3 o T3

PAGE 81

65 O O o in o o 8 o o o o o o O) (£> CO < q: hin q: o r> Io o o ro \ndo

PAGE 82

66 cells, expressing antigens of the F host to which the BIO. BR lymphocytes were sensitized. No proliferative activity was directed against the syngeneic donor cells which initiated the GVH, namely BIO. BR (H-2^) This indicates that the host's lymphocytes are not proliferating against the donor lymphocytes, as in the nonirradiated F, animals (see Table 3). Nor is reactivity seen against B10.D2, an unrelated third party haplotype (H-2^). Some cross reactivity of the response of the primed cells is seen on day 1, but this is not considered significant because of the loss of activity on day 2. The optimal peak of secondary 3 H-TdR incorporation of either MLR or GVH primed cells is always found on day 2. Thus, the cells do not seem to be capable of reacting towards antigens which are different from those antigens which initially triggered the GVHR; and this makes this reaction appear to be similiar to the in v^tro primed lymphocyte test (PLT). In addition, GVH activated lymphocytes do not react towards the T cell mitogens, concanavalin A (Con A) and phytohemagglutinin (PHA), or lipopolysaccharide (LPS), a potent murine B cell mitogen (Table 5). (BIO X B10.BR)F^ hybrid mice reconstituted with parental BIO. BR spleen cells normally died between days 10 to 14. Cell populations recovered from the spleens and livers of mice showing signs of severe GVHD on day 8 and 9

PAGE 83

67 Table 5. Inability of GVH primed cells to respond to mitogens. H-TdR incorporation CPM+SD (a). addition normal GVH normal cells normal cells cells primed treated with treated with (b) cells anti-Ohy+C' anti-I-A+C' (c) (d) (e) none 4668+880 1349+12 6303+ 1477+ 163 Ccn A(f ) 54043+6244 417+94 2700+595 22311+1345 LPS (g) 98977+5401 2531+81 71110+655 19737+ 517 a) MLR assays were harvested at 48 hrs following a 12 hr pulse with 1 nCi H-TdR. b) Spleen cells from a normal healthy mouse (BIO. BR) were used in this study. 5 x 10 cells were placed in each well of a microtiter plate. c) (BIO X B10.BR)F, mice were sublethally irradiated and reconstituted with 40 x 10 BlO.AQR splenocytes. The mice were sacrificed on day 5 after reconstitution. d) Normal splenocytes were treated with anti-Thy 1.2 antibody and oanplement immediately tefore culturing the cells. e) Normal splenocytes were treated with anti-I-A antibody and complement immediately before culturing the cells. f ) Concanavalin A dose was 5 |ig/ml. g) Lipopolysaccharide dose was 100 |ag/ml.

PAGE 84

"^wsipriwr"68 after engraftment were compared in PLT with in vitro activated BIO. BR anti-(B10 x B10.BR)FPLT cells. As presented in Table 6, major differences exist in the response pattern of the in vivo and in vitro activated cell populations. Cells obtained from either the spleens of livers of the F, hybrids undergoing GVH were restimulated by cells from strains carrying the H-2 haplotype. However, stimulation with cells from recombinant strains possessing either the H-2D or H-2K antigens, e.g. BIO.MBR, B10.A(2R) produced little if any reactivation of proliferation, and no cross reactivity was observed on unrelated third party strains. In contrast, MLR generated PLT cells exhibited strong secondary responses to strains carrying the H-2 haplotype as well as H-2 haplotype derived K and D region antigens. In addition, cross reactivity on unrelated third party strains was observed such as BIO .T( 6R) While a number of factors may account for the different patterns of reactivity exhibited by in vivo and in vitro primed cells, the most likely one is the protocol used herein to obtain responding PLT cell populations from organs undergoing GVH reactivity. Overnight incubations of the in vivo primed cells in the presence of cell debris and dying PMNs could alter markedly the reactive patterns. To determine the effect of this procedure, cell mixing

PAGE 85

69 ^ 2 w 8 VD H EC •H O PQ Cn c CO 3 c 8 (0 Q CO. 6 c o 4J § I n O u r3 -H ^ C (1) (1) 2 > cn o I Q S IS cn CO CM i' CO CM in CM + 1 CM '3' n n CM ^1 in o CN O 00 O O 00 S' O CN CM O n + 1 +1 r^ CN rH C3> CM m CT\ 00 XI XI s Q O Cm O iH o fH PQ rH M X) X! XI x> XI CO in o CM 00 00 00 CO ^ CM in o + 1 +1 +1 +1 +1 +1 00 ON rH CO rO r00 iH iH in in CN rH rH ro in VD CO rH ^ H ^ CN rr r•^ rH vD CM r~kO rCN CM ^ vo i-i, iH in iH + 1 + + +1 +1 +1 CM CJN r~ -H O CN ^D CN VD cn i-i 00 •^ CTi O rH CTi rH CO CM CM r^ U3 TJ< CO CO CT\ o o CN "a+1 +1 +1 +1 CTi CTi CT> a> VD CTi 00 VO r r~o CTi o in cTi in CM ^ i-l CM r in CTi O rH CN +1 +1 00 CM o CO I— I \D 00 rVO 00 Xi CJ< X3 X3 TJ T) M M X> Xi M J^ X M XI JQ X> J4 M M M '^ JH Xi JJ XJ Jk! T) XI XI ^ 2 < 8 < < 05 a CO in CN. CM rH n CO CTi CN rH CO CM in 00 o o cr cc; o o o o o o rH rH rH nH rH rH to cQ OQ ffl cq m o a IW 4J O (0 •H H TJ ^2 4-) +J (1) T! rH Jj •H rH OJ (1) (31 & C O o ^s en • m 00 O (1) o; tn +J -H cn cn (U c ^8 5g .Sl x: jJ o 3 C ^S 8 -S S rO XI • u o d) o S ^5 Ti 0) u (U w o (0 > y •H -H Ci -O rH (0 -U •H (0 -H c >i cn )H rH C 0) rH (U > m (1) O • X> fl) 3 'U c ^ 8 •H 00 00 73 UH 3 cn cn c O O i cn cn c ?S58. c E>, cn rH JJ O CO Dj-h c cn > 0) rH H rH rH Cn u m 8 • • o^ O 33 O C CQ (3 CQ cn o 0)

PAGE 86

70 experiments were performed. Responding cell populations primed in the MLR of BIO. BR responding against (BIO x B10.BR)F, were mixed with equal numbers of cells removed from the spleens of (BIO x B10.BR)F^ mice undergoing GVH reactivity following reconstitution with BIO. BR spleen cells. Following an overnight incubation, this mixed cell population was tested in PLT and its reactivity compared to the responses of the two individual cell populations. As shown in Table 7, Column 3 the mixed cell population exhibited the reactivity of the in vitro primed cell population (Column 1). In a more direct approach to determine if differences arise due to culture artifacts, cells obtained from spleens undergoing severe GVHD separated on Ficoll-Hypaque density gradients (d=1.101) were examined directly. This procedure results in a responding purified population contaminated with less than 2% PMNs As shown in Table 7, Column 4, the responding cell population obtained in this manner still exhibited a PLT reactivity identical to the in vivo primed cells which had been incubated overnight. Thus, the differences in the reactivity of the in vivo versus in vitro primed cells does not appear to be dependent on the handling of the cells. K and D antigens are known to exist on every cell of the body, while I region antigens are found on only a

PAGE 87

71 w 0) w c a 8 •rH ro M-l CO 8 (0 e c o > I W 8 + = (0 o H 0} I T3 CO 8 o TJ m V-l 0) -o CO H-8 Q 8 CO en en ^ 00 OJ ro ro o VO rH <£> ^ UD rH. iH ^ + + + o o o rsj in in cx) CN riH CO n rH 00 r~O "T Vr> rH O r-H 00 ^ rO ID VD rO ^ rH CN rH rH + 1 + 1 + 1 + 1 + 1 + 1 r~ 00 00 o o 00 ro 00 VD r^ ri r~ O vD CM CO CN 00 00 u> 00 vjD in rH ^ "* 00 vo 00 •^ m ^ 00 +1 +I+I+I+I+I+I r-i rH ID in 00 CN O in 00 'a Ti crx> jQ M J3 Jb^ J! J^ jQ -r-i M XI X> M M Xi -1-^ M JH Xi Xi M M-t-\ M Xi £> St £t M -r^ 00 in ci '* < CO x: C G CO
PAGE 88

72 selected set of cells such as B cells, monocytes, skin dendritic cells and liver Kupffer cells. The possibility exists that the cells which react to the K and D antigens are filtered out before they home to the spleen or liver. To exclude this possibility, embryonic B6 mouse fibroblast monolayers were established and passaged three times. These cells possess the K and D antigens, and do not express the I-A antigen. In vitro primed BIO. BR anti-(BlO X B10.BR)F, cells were then incubated for 2 hours on this monolayer and then gently rocked off the monolayer. Approximately one half of the primed cells were removed by this treatment. These cells as well as a sample of the original primed population were tested in PLT. The results shown in Table 8 show that both populations were still capable of responding to the K and D antigens found on BlO.MBR and BIO.GD. Although cytotoxic T cell activity was not tested before and after adsorption, it appears that this filtering mechanism does not occur in vivo because cytotoxic T cells are found in vivo in the spleen and liver and that they do respond to the K and D antigens (see below). Treatment of the in vivo primed BIO. BR anti-(BlO x B10.BR)F, cell populations with anti-Thy 1.2 antibody plus complement totally abolishes the PLT reactivity (Table 9). k Treatment with monoclonal anti-I-A antibody plus

PAGE 89

Table 8. Absorption of MLR primed lynphocytes fails to remove proliferative responses towards K/D antigens. 73 H2 Genetics H-TdR Incorporation CPlVH-a) (a) Stimulator Trea1 none mient: (b) strain K A J E D absorption none 8236+1126 8478+2284 BIO. BR k k k k k 11043+1766 9187+ 184 B6 b b b b b 88425+ 648 69299+6033 B10.A(3R) b b b k d 99479+3454 83931+1179 BIO.MER b k k k q 15213+2087 10245+2835 BIO.GD d d b b b 52650+ 47223+6376 B10.A(4R) k k b b b 15215+ 42 15057+1344 a) b) PLTs were harvested at 48 hrs following a 12 hr pulse with 1 HCi of H-iaR. Primary MLR consisted of BIO. BR splenocytes incubated with BIO splenocytes for 5 days. An aliqout of primed cells was incubated for 2 hrs, v\^ile the raraining cells were incubated over the B6 fibroblast monolayer for 2 hrs. After the incubation the cells were gently rocked from the plate and the lyn^oblasts were recovered. The fibroblast monolayer was nade by taking B6 anbryoes and mincing them into single cell suspensions using a 0.5% trypsin solution. The fibroblast monolayer was passaged three times during a period of 18 days.

PAGE 90

74 Table 9. The PLT activity of GVH primed cells after treatment with various monoclonal antibodies plus conplement. H-'MR incorporation CPM +SD (a) stimulator cells primed cells treated with C (b) none BIO BIO. BR (BlOxBlO.BR)F B10.A(3R) B10.A(4R) B10.A(5R) 604+382 6409+175 649+ 69 6361+239 5478+ 59 623+ 45 3320+494 primed cells treated with anti-Thy + C (c) 1574+544 1490+ 37 1762+ 50 1701+ 33 1619+ 67 1152+183 1128+ 79 primed cells treated with anti-I-A + C (d) 1778+266 16711+880 2252+ 47 17570+544 14862+145 3360+771 13892+951 a) GVH reactivity vas induced in sublethally irradiated (BIO x B10.ER)F^ mice with BIO. BR splenocytes. Eight days later the spleens of these animals were ronoved and prepared into a single ^11 suspension and used. The cultures, consisting of 30 X 10 cells per well, were pulsed with 1 |aCi of H-TdR for 8 hrs on day 2 of the reaction. b) Spleen cells recovered from the GVH reaction were treated with rabbit complanent alone. c) Spleen cells recovered from the GVH reaction were treated with anti-Thy antibody plus complement immediately before culturing. d) Spleen cglls recovered from the GVH reaction were treated with anti-I-A antibody plus complement immediately before culturing.

PAGE 91

75 complement (kills 60% BIO. BR splenocytes and 5% BIO splenocytes) a procedure which enriches for T lymphocytes by depleting the 30 to 40% contaminating B lymphocytes, increased responsiveness in those combinations already exhibiting a positive response, but did not produce reactivity against strains which showed negative responses. C57BL6 anti-BALB/c reaction Irradiated BALB/c mice reconstituted with B6 spleen cells normally died by day 9 to 10 after engraftment. The pathology was identical to that found in the BIO. BR anti(BIO X B10.BR)F^ combination. Cell populations recovered from the spleens, livers and lymph nodes on day 6 were tested in PLT for their ability to mount a secondary proliferative response. As can be seen in Table 10, all three populations exhibited strong proliferative responses against cells from mice possessing the H-2 haplotype, e.g. BALB/c and DBA/2 or from mice possessing the H-2K and I-A regions, e.g. BIO.GD. Little, if any significant reactivity was elicited against third party strains including those expressing H-2D region antigens, e.g. BIO.AOR), B10.A(5R) and B10.T(6R). Thus, these alloactivated cells appear to recognize primarily determinants encoded by genes located on the left side of the H-2 complex.

PAGE 92

76 i 6 c o E M CO 8 8 w be is IS W CO o o o iH CN en in rH lo vo rH in ^ 00 VD m en o in in n rH Of) en >x> r-i fTl +1+ +1+ r^ m CN CN x) * in o in s VO •H Q) I — ( v^ m XI o

PAGE 93

77 Secondary responses of B6 anti-BALB/c PLT primed cells primed in MLR are also included in Table 10 for comparison. These cells exhibited a quite different pattern of reactivity from GVH primed cells. Strongest responses were elicited against strains possessing the H-2 haplotype or the H-2K and I-A regions; however, strong responses also occurred against strains possessing the H-2D region antigens. In addition, cross reactivity against a number of unrelated third party strains, e.g. B10.A(2R), SWR and AKR, was observed. The BIO.RIII anti-BlO .A( 5R) reaction In another whole H-2 disparate GVH, BIO.RIII antiB10.A(5R), similar results were found (Table 11) like those in the two previous combinations. The MLR generated primed cells respond towards the K and I-A antigens present on B10.A(5R) and B6 mice. Strong stimulation was also found on mouse cells possessing the K antigen BIO.MBR as well k d as the I-E and D antigens, e.g. BIO. A, BIO.AQR, BIO.TL, BIO.SOR), A.TFR5, B10.M(17R) and BIO. BR. In contrast, GVH primed cells only recognized the B10.A(5R) and B6 mouse cells. The lack of a response to the BIO.MBR cells apparently eliminates stimulation due to the K antigen.

PAGE 94

Table 11. Canparison of PLT of BIO.RIII anti-BlO.A(5R) cells generated either in MLR or in GVH. 78 HK~ 2 Genetics A J E D ^H-TdR Incorporation CEMfa) (a) Stimulator Strains In vitro primed cells (b) In vivo primed cells (c) none 3832+ 2227 489+ 257 BIO .RIII r r r r r 4229+ 858 151+ 23 B10.A(5R) b b d d d 85395+ 1382 30006+ 168 B6 b b b b b 93669+ 1283 26426+4045 B10.D2 d d d d d 51594+ 7504 5886+1625 BIO .AQR q k k k d 66655+ 2270 2873+1453 BIO .A k k k k d 57325+ 49 1704+ BIO.TL s k k k d 52795+15596 2814+1071 BIO.SOR) s s k k d 28573+ 3896 468+ 251 A.TFR5 f f k d 32065+ 8188 374+ 173 BIO.MER b k k k q 43565+ 1493 2398+ 155 B10.M(17R) k k k k f 27360+ 2955 1533+1327 BIO. BR k k k k k 20357+ 3594 3608+ 91 B10.A(4R) k k b b b 19194+ 1996 610+ 1 BIO.GD d d b b b 21432+ 2476 997+ 250 BIO.WB J J j j b 8476+ 888+ 347 B10.M f f f f f 11401+ 2331 1148+ 286 a) ELTs were harvested at 48 hrs following a 12 hr pulse with 1 |jCi of H-TdR. b) Primary iMLR consisted of BIO.RIII splenocytes incubated with irradiated B10.A(5R) splenocytes for 6 days. c) GVH reactivity v^as induced in sublethally irradiated B10.A(5R) mice with BIO.RIII splenocytes. On day 5, the spleens were collected, made into a single cell suspension and cultured overnight prior to use.

PAGE 95

79 Again, the GVH primed cells display a more restricted response pattern than do primed cells established in vitro. The BIO.GD anti-BlO .M(17R) and BIO .WB anti-BlO .M(17R) reactions Two other genetic combinations with H-2 disparate regions studied were BIO.GD anti-BlO .M( 17R) and BlO.WB antiBlO.MdVR) as shown in Table 12. Both in vitro primed cell populations responded towards cells from mice which k k f possessed either the K ,I-A or D molecules. In contrast, the GVH primed cells reacted only to those cells which possessed the I-A molecule, e.g. BIO. BR, B10.M(17R), BIO.MBR, BlO.AQR and B10.A(4R). Although minor • k f reactivity is seen against E and D molecules, the GVH primed cells clearly did not respond to the same magnitude as the in vitro primed cells do. Even though the BIO.GD anti-BlO .M( 17R) and BlO.WB antiB10.M(17R) reactions are directed against the B10.M(17R) haplotype, these two in vitro reactions are not identical. The BIO.GD anti-BlO .M( 17R) primed cells recognized some determinants which the BlO.WB anti-BlO .M( 17R) primed cells did not, such as those found on the BIO.F cells. What these determinants are is not readily distinguished from the present data. However, this antigenic determinant is readily detected by the in vivo primed cells giving some

PAGE 96

80 § ^, i' c o H I i I I •H JJ --^ O 3 r— iH re § S gj • • S o o c 2 (0 r^ 8 S • • o o ^ 4J -^ XI c a; rH t3 ffl -•H O O 1-1 rH rH CXCQ CQ g 05 8 2 • • o o rH r-\ m QQ Q w O xi CO w ronr^vOTj"rHPoinrHVDvooo CNOrH'3'oorsioornooor~-m ronrHcriCT>ro(T>oinnrH + 1 + 1 + 1 + 1 + 1 + 1 + 1+1 + 1 + 1 + 1 + 1 I CTN 00 ^ ^ r-in (T\ (N >* o IT) Tj r~^VDVDrHTj* a^ o^ in n rH rH + 1 + 1 CM CM in vD in o vovDrH'^rHinr^cM'^ CO o^r^rfincMinorn n lna^cNro^~^or~lnrH oo ,rH. CM CM en r-\ + + + +I+I+I+I+I+I I +1 n tj. in CM CM o ro o r-'3'O'^ o cNrrvo'^n-^r'^CM'^ a\ CM CM CM '51' ro rH rH r~ CTl 00 VD CTi CM m H +I+I+I I ^ in c^ Lnoinooo rnoo ^in"ti'cricx3r~^ rHo^ 'S'Ln'TrOrHrHro CN I ja 73 XI HH ^ CTTJ j3 TJ iM )H m ana I X! ^O -r-iJ*; Jk!Js-i CO ft cr rH 05^^ CTl S 0:5 W fa in 8 Q i S S^ i S* < W S S '::^:;;id'ri'i^'z^'-''-''-''-<'-irHrHrH CCQCQOQCQCQOQCQcacQOQtricQOam 0) M W" S-i M O >1-r( 0) -rH CN O rH '-^ VD rH JJ 8(0 in x; • O -U >i-H >H O HH 0) CU rH ^ +J CQ CO ^ 0) D Vj 4J CO c Sh •a CO JJ O -H •s si CO OJ -H rH TJ 4-> r^ Q. Q) CO rH CO Er S v f c J T3 -H CO S (T3 Eh -H s-i an o r-j jj M M > rH rn S PU -H O OQ O C rH JJ O -rt (0 J3 O

PAGE 97

81 possibility that these cross reactions could either be due to cross reactivity to the I-A^ molecule which is common to the I-A molecule or that the I-E^ cross reacts with ]^ the I-E molecule. The latter explanation is favored because the BlO.GD does not express the I-E molecule and is therefore capable of recognizing the entire I-E molecule as foreign, while the BIO.WB animal does express the I-E molecule and does not have to respond to an entire I-E molecule as would the BlO.GD cells. The B10.M(17R) anti-BlO.GD and BIO.WB anti-BlO.GD reactions When another set of whole H-2 reactions is established against the BlO.GD mouse [B10.M(17R) anti-BlO.GD and BIO.WB anti-BlO.GD] a different pattern of data is seen (Table 13). The predominate in vitro stimulatory antigens come from the left side of the H-2 complex, i.e. K and I-A as exhibited by BlO.GD and B10.D2. Reactivity is also observed on B6, BIO.AOR) and B10.A(4R) cells which possess the I-J and D antigens. In addition, cross reactivity on the I-A molecule is also postulated for the B6 and BIO.AOR) cells, but these responses to the antigens do not seem to be additive when compared to the B10.A(4R) response, This cross reactivity on I-A could be due to the shared antigens Ia8 and Ial5 which I-A has in common with I-A to which the I-A molecule lacks. In addition, the GVH

PAGE 98

82 b ^ t m • n r-l 8 fl) £L> rH > 1 § 8 'f o CM c 8 & c (0 8 I t3 O Q) rH s m -H a •H > c 8 ^ S 8 rH O 03 C to Xi — 69 o •a o 4J o rH CQ ^ CO o •r-l I 5 8 r• nH O ^ rH S CQ • I O -rH rH 4-) CQ C (0 Q W H O en CO vDvr)OvDrHrooor-~r~ooo (Ti-* cN'^cNinininr^'^CNirHiH a\rCNrH.iHinoOCN (NCN t-i + 1+1+1+1+1+1+1+1+1+1+1 1 +1+1" I^VDo*ocNlna^c^JVDO cnvo rHcTiinoOrHrovDinr^'* rofsi Oi (Ti rH CM in A" o + 1 CM in in CN + 1 I CN VD CM ro r^ "Scn in rH 00 rH m ro Tj" -^ rH r-{ 00 rH rH i-H vo CM rcvj n ro •* CM ro in o CM CM rH m cm i S-S D O VD rH rH rH CQ IH O O MH Sh KH CO is CQ MH O 00 o <0 TJ rH 0) CO jj -ri m CO CO S ess CO S (0 Eh -H Vj !^ Jt^ ^ P4 dl -rH (0 XI U (D rH CO O CQ C 4J TJ (1) >H CD rH O in c >^ ^ ^ rH >1 0) 2 'S O iB •H 0^3 TJ C C rH ••H --^rH 4j ck; fl) o r^ o (C rH H SrH § S-S O CQCO O

PAGE 99

83 primed cells also appear to recognize these antigens, since GVH primed cells apparently only respond to I region antigens. Thus, this cross reactivity is likely to be due to class I antigens. Interestingly, the in vitro B10.M(17R) anti-BlO.GD reaction appears to detect a cross reaction on the K^ molecule, since BIO.TL, BIO.S, BIO.SOR) and BIO.HTT cells induce stimulation by the primed cells. In contrast, the GVH primed cells do not appear to recognize this cross reactivity; again the response seen by the GVH primed cells is different from those of the MLR generated cells. These cross reactions to K^ which are seen by the in vitro primed B10.M(17R) anti-BlO.GD cells are not seen by the in vitro primed BIO.WB anti-BlO.GD cells. The I region GVH reactions A number of mouse strain combinations exists with limited I region disparity. Two of the most extensively studied combinations are B10.A(4R) responding against B10.A(2R), two congenic lines differing genetically between the H-2 I-A and D, and BIO.AQR responding against B10.T(6R), two congenic lines differing genetically throughout the I regions. These two combinations are considered to represent, respectively, an anti-I-E^ and anti-I-A^ response. The secondary MLR responses of B10.A(4R) anti-BlO .A( 2R) and

PAGE 100

84 r BIO.AQR anti-(B10.T(6R) x B10.AQR)F PLT obtained from 10 day GVH spleens or primary MLR are examined in Table 14. As presented in Table 14, the PLT cells generated in MLR exhibited strong responses against the primary stimulating strains as well as strong cross reactivity on third party strains. For example, in the B10.A(4R) antiB10.A(2R) reaction the specific antigen is I-Ea^6^,yet k s BIO.HTT which has I-Eq q is recognized by the in vitro primed cells as well as the BIO.RIII cells. Likewise, a great deal of cross reactivity is seen by the BIO.AQR anti-(B10.T(6R) x B10.AQR)F^ reaction, a reaction supposedly directed at the I-A^ molecule. Mouse cells from BIO, B10.S(7R), BIO.M and BIO.RIII mice elicit a response by these primed cells. Again as was demonstrated with the other reactions the GVH primed cells demonstrate marked specificity. In the B10.A(4R) anti-BlO .A( 2R) reaction the GVH primed cells only respond to the specific antigen: k k I-E a 6 found on BIO.AQR, B10.A(2R) and BIO. A. While in the BIO.AQR anti-BlO .T( 6R) reaction only those cells possessing the I-a'^ molecule found on B10.T(6R) and DBA/1 cause restimulation. K/D region GVH reactions Two strains namely BlO.MBR and A.TL possess genetic differences at only K and D loci. The MLR and GVH reactivity

PAGE 101

85 ^ b V •H 0) 2 to 8 • ^ tJ •H > 1 •H c M H a o ^^ M 4-> T5 H > g •rH C U H CU V u u •H I t3 Q W < |.s CO CO o n (T\ CO ro o^ r~ ^ 00 a> CN iH M 00 4? OJ + 1 + 1 + 1 + 1 1 iH 00 on in VD ID .H CM CM iH QOOOfN rHint^-OVD (TiOOO I +1+1 +1 +1+1 I +1 +1 +1 r~ 00 (— I t^ r~cvj f\j VD fHovofNfH ^^J^^r^ ("2 r-i (J\ r-i fT) cNino iH CN iH CO in in in 00 iH r-l CM CM CM rH rVD 00 'i' CM +1 +1 +1 +1 CO ro' 00 ^ 4J >-l H g S4'0 en w C rH B8 H o c M -H tn c 00 (1) s t^ o p O T3 ill • a) 0) O M C OQ --H W CO TJ 0) C a) 4-5 m !:i f^ "* (COW -a c w (0 0) •H CV m ro • C7I JS O C 4-) -H d) P3 C PS •iH ^ 0) O c ^ OQ CQ PQ DQ 4J "O en (0 (1) >i Ti c ^ Q) (fl 4J 4J CO CO c ^ 8 ^ to to (U 2 4J c S CU S' -a < 8o^ 3 rH .. tJ OQ 73 c a) •^ jC 4J g SrH >,88 •H rHm (fl eg CO P Qi C E? 5 ':d • > rH CLi JH u OQ to a V-l o (0 xt o

PAGE 102

86 of these combinations have been examined. Results from the reaction of BlO.MBR responding against (BlO.MBR x A.TDF, hybrids, which involves a haplotype mismatch at both the K and D, is presented in Table 15. PLT cells generated in vitro exhibited strong secondary responses against third s d party strains expressing K /D antigens, syngeneic with the primary stimulating antigens. No cross reactivity was observed with other haplotype products. Such specific reactivity against class I products is well documented. In contrast, it has not been possible to obtain primed cells from organs undergoing GVH in K/D disparate combinations s d which will proliferate against K or D region antigens in PLT, despite the fact that the mice are dying from GVHD. Another class I mismatched reaction was explored on a gross mortality level, namely the B6 anti-B6 reaction and the B6 anti-B6 reaction. This combination represents a mismatch directed at a variant K molecule. The K molecule found on B6 cells represents a K molecule with 2 amino acid substitutions (Wakeland, personal communication). The reactions observed proved to be one way reactions: one half of the B6 mice reconstituted with B6 cells died within the first 15 days of the reaction, while none of the B6 mice reconstituted with the B6 cells succumbed to death. Thus, a variant molecule is

PAGE 103

l^le 15. Canparison of PLT of BIO.MBR anti-(A.TL x B10.MBR)F, cells generated either in MLR or in GVH. 87 i'H-TdR Incorporation CPM+SD (a) t stimulator Strains H2 Genetics In vitro primed In cells (b) vivo primed K A J E D cells (c) BIO.MBR b k k k q 1921+ 37 397+ 57 (A.TLXB10.MBR) b k k k q ^1 s k k k d 21241+2715 264+ 15 BIO.TL s k k k d 28455+2710 203+ 40 B10.S(7R) s s s s d 26055+ 121 BIO.SOR) s s k k d 20570+1066 257+ 7 BIO .HTT s s s k d 24706+ 610 258+ 35 B10.T(6R) q q q q d 21645+ 227 240+ 57 B10.D2 d d d d d 27624+1587 184+ 4 BIO b b b b b 4597+ 264 135+132 BIO. BR k k k k k 4661+ 184 365+100 BIO.GD d d b b b 5653+ 938 243+ 25 a) PLTs were harvested at 48 hrs following a 12 pulse with 1 piCi of H-TdR. b) Primary MLR consisted of BIO.MBR splenocytes incubated with irradiated (A.TL x B10.MER)F, splenocytes for 7 days. c) GVH reactivity was induced in sublethally irradiated (A.TL x B10.MER)F, mice with BIO.MBR splenocytes. On day 6, the spleens were collected, itade into a single cell suspension and cultured overnight prior to use.

PAGE 104

88 sufficient to induce lethal GVHD, but for some unknown reason the individual possessing the variant molecule is not capable of producing a lethal reaction in those animals which possess the parent molecule. Minor histocompatibility GVH reactions In the in vitro MLR system there exists another antigenic system which is known to cause strong T lymphocyte proliferation similar to I region antigens. This system has been termed Mis. The Mis antigen system was originally described by Festenstein (97, 98, 99) by using the primary MLR on H-2 identical mice. The original reaction came from the combination BALB/c anti-DBA/2. Later by using the F^ tests and PLT assays other various mouse strains have been given Mis designations. BALB/c (Mis ) cells which express the silent Mis allele are able to respond against cells which have the Mis antigen such as is found on DBA/2 cells. Indeed, in vj-tro primed cells appear to be able to proliferate towards mouse strains which apparently are Mls^ positive, i.e. DBA/2, Dl.C, AKR and possibly SEA (Table 16). In addition, positive reactions are also seen on Mls'^ positive cells: CBA/J and RF. Likewise, Festenstein has given the mice DBA/2, Dl.C, NZB, SM, DBA/1 and AKR the designation Mls^, while CBA/J and RF are Mls'^ (101) and these two

PAGE 105

Table 16. Canparison of PLT of BALB/c an ti -DBA/2 cells generated in either MLR or in GVH. 89 H-2 Mis H-TdR Incorporation CIMfSD (a) Stimulator strain In vitro primed (b) In vivo primed (c) none 309+ 8 1819+684 BALB/c d b 10615+ 496 1610+180 SBC d b 42867+3975 1713+ 25 SEA d a 90114+2165 1905+373 DBA/2 d a 80777+2568 5390+231 Dl.C d a 92882+3137 3438+484 BIO. BR k b 1735+412 CBA/Ca k b 2072+911 RF k d 83890+2890 — CBA/J k d 3534+ 54 C3H/He k c 34850+1975 1934+ 23 a) PLTs were harvested at 48 hrs following a 12 hr pulse with 1 HCi of H-TdR. b) Priitary MLR consisted of BALB/c splenocytes incubated with irradiated DBA/2 splenocytes for 6 days. c) GVH reactivity v,qs induced in sublethally irradiated DBA/2 mice with BALB/c splenocytes. On day 5, the spleens were collected, nade into a single cell suspension and cultured overnight prior to use.

PAGE 106

90 designations cross react strongly and are possibly the same antigen (100, 101). However, since SEA induced strong secondary proliferation, its Mis phenotype is assumed to be either Mis Similar findings are exhibited by B10.D2 anti-DBA/2 primed cells (Table 18 column 3). As is the case found in I region GVH, Mis disparate GVHs produced very few lymphoblasts in the spleen but an increased amount of PMNs frequently interferes with PLT assays. However, GVH primed cells do appear to proliferate when Mis or Mis positive cells, e.g. DBA/2, Dl.C, SEA and (RF or CBA/J) are cultured with the primed cells (Table 16 Column 2) Another Mis disparate GVH was established using CBA/Ca (Mis ) cells reacting against AKR (Mls^) cells. The results are shown in Table 17. In this experiment positive results are seen by proliferation responses against AKR, DBA/2 and SEA cells. Again this reaction shows weak responses directed against the Mls^ determinant in contrast to the in vitro reactions where strong responses are demonstrated. Whereas the in vivo primed cells failed to yield large numbers of primed cells towards minor histocompatibility antigens such as Mis, it was possible to examine and define Mis antigens carried by various mouse strains by utilizing in vitro MLR primed cells. From the

PAGE 107

91 Table 17. Canparison of PLT responses of CBA/Ca anti-AKR generated in either MLR or in GVH. H-2 Mis H-TdR incorporation CPM +SD (a) Stimulator In vitro primed In vivo primed cells (b) (c) none 4919+ 152 1317+ 199 CBA/Ca k b 4137+ 311 1381+ 227 BIO. BR k b 3315+ 365 1990+ 75 AKR k a 89997+2524 6811+ 72 C3H/He k c 21202+ 53 1385+ 349 BALB/c d b 44178+7094 3535+ 116 B10.D2 d b 20362+ 113 4488+ 226 SBC d b 23793+ 714 4832+ 503 DBA/2 d a 97441+2568 10155+1014 SEA d a 67067+1496 7502+1318 a) PLTs were harvested at 48 hrs following a 12 hr pulse with 1 |iCi of H-TdR. b) Primary MLR consisted of CBA/Ca splenocytes incubated with irradiated AKR splenocytes. PLT cells were collected on day 5. c) Sublethally irradiated AKR mice received CBA/Ca splenocytes. At 9 days after reconstitution, the spleens were collected, prepared into a single cell suspension and cultured overnight prior to use.

PAGE 108

92 (1) 00 00 00 CN vo I I I I I I m CNJ CO + 1 CTi s rrH o iH O 00 iH CO 00 + 1+1+1 1 •* 00 o ro CO ro in r~ r~ O vo 00 o r-•^ rr + I + I + I I n vo 00 00 rH 00 ro in vo rH in cTi <* ^ o r+ + vo CN vo CM m in m in cr> o in ^ in ^ CM 00 r^ rH o^ vo CM in •^ CM vo CO o O CM +I+I+I+I+I+I+I t^ •^ rH in ^ o CO ro 00 in r~ tTv CM 00 in !• 00 r^ 'T vo '0 M M M J^ ^ M M CM Q • .2^ ^ g ^ O CQ < i < < p i CO rtc rH tS] p 03 Z C 88^ m S i IJ4 in in w OO 2 S O U U £-• ^ & s ^ •H ^ C • ,cn UH 0) rH VO cn a i w cn CO "S" -Sc 1^ a CO -O -O 0) 0) M-l 0) (U rH e rH X) p 5 CO Q) (1) •H J-l 8 CO u ^ Eh CC rH ^ l8 to

PAGE 109

93^ reactions described here, BALB/c anti-DBA/2 (Table 16) and CBA/Ca anti-AKR (Table 17), PLT analysis have identified DBA/2, Dl.C, RF, AKR, and SEA as being Mis similar. When further experimentation was performed on the genetics of the Mis, it was noted that this system is not as simple as it was inititially described. For example, when F^ hybrid mice were constructed [(B10.D2 x AKR)F and (BALB/c X AKR)F^] so that the Mls^ antigen was blocked out by the AKR background, these F^ mice demonstrated strong proliferation against DBA/2 cells. This would indicate that another antigen is capable of causing in vitro proliferation other than Mis and H-2. Further studies have shown that SEA and NZB probably lack this antigen or antigens because they also are capable of recognizing DBA/2 and Dl.C cells in PLT assays (Table 18 columns 4 and 5). Additional experiments have shown that different strains of mice may recognize minor histocompatibility antigens in the context of the H-2 antigens. For example, in Table 19 the BIO. PL anti-PL and BlO.SM anti-SM combinations produce good primary reactions only in one direction and the primed cells respond in PLT assays. From PLT assays, it appears that the Mis '^ antigens are responsible for some of the stimulation which is observed because the primed cells respond towards cells from DBA/2, SEA, Dl.C, AKR and NZB.

PAGE 110

94 Table 19. Surtmary of minor histoccmpatibility in vitro assays in inbred H-2 mice. Reaction: H-2 Priitary Reaction (a) PLT (b) BIO. PL anti-PL u PL anti-BlO.Hi u B10.3y[ anti-a4 V 3A anti-BlO.SM V BIO.S anti-SJL s SJL anti-BlO.S s C3H.^ anti-BlO.F p BIO.F anti-C3H.ISB p SWR anti-BlO.Q q SWR anti-DBAA q BIO.Q anti-SWR q BIO.Q anti-DBAA q (BlO.BRxSWR)F anti-SWR q BALB/c anti-DBA/2 d B10.D2 anti-DBA/2 d SEA anti-DBA/2 d SEC anti-DBA/2 d NZB anti-DBA/2 d AKR anti-CBA/J k AKR anti-C58 k CBA/J anti-C58 k CBA/J anti-AKR k C58 anti-CBA/J k C58 anti-AKR k AKR anti-RF k CBA/Ca anti-AKR k + + + + + + + + + + + + + + + + + + + a) The ability to generate a positive primary MLR is demonstrated by either + or -. b) The ability of the primed lyn^ocytes to respond and give a specific PLT is detionstrated by either + or -.

PAGE 111

95 a k Using mice carrying the H-2^ and H-2 haplotype strange reactions are found. In the H-2^ system, SWR reacts strongly to DBA/1 and BlO.Q, BlO.Q reacts strongly to SWR and DBA/1, while (BIO. BR x SWR)F reacts strongly to SWR. Despite these strong primary reactions with large numbers of blast cells, no specific PLT reactivity has been obtained to date. These primed cells recognize self as strongly as they recognize the original stimulating cells. Likewise in the H-2 system certain combinations yielded good primary MLRs, yet no specific PLT activity was obtained. The possibility exists that the primary reaction was still continuing when the primed cells were taken and this would explain why no specific pattern was detected. However, this explanation does not appear likely, since these primed cells could be extracted 2 weeks after the culture was established and the cells have had sufficient time to revert back to small lymphocytes. Again these cells failed to show any specific proliferation. Another combination yeilding interpretable results was SJL reacting against BIO.S. Here the primed cells reacted against BIO.S, CE and C58, whereas AKR, DBA/2, SM, PL and CBA/J cells did not induce any proliferation by these primed cells. Thus, it appeared there may be another minor histocompatibilty reaction which did not appear to involve the Mis system.

PAGE 112

96 ii. Cell mediated lympholysis assays Results of the PLT tests in entire H-2 mismatched combinations raised the question of whether any reactivity against K/D antigens in vivo could be found. For this reason cells recovered from organs undergoing GVH in each combination were tested for cytolytic activity in CML against targets expressing K/D (or I) antigens syngeneic with the primary stimulating cells. BIO. BR anti-(B10 x B10.BR)F reaction Results presented in Table 20 using the BIO. BR anti(BIO X B10.BR)Fj^ system indicate that killer cells can be recovered from the spleen on day 5 of the GVH reaction. MLR generated cells lysed cells which express either the K*^ or D e.g. BIO.MBR and B10.A(2R). Cells recovered from the GVH spleen were capable of similar patterns of killing as exhibited by the MLR generated killer cells. If the cells were cultured overnight to get rid of the PMNs, no killing activity was exhibited. However, if these cells were cultured overnight with EL-4 derived interleukin 2, killing activity was retained, but restricted to the cells which possess the K antigen.

PAGE 113

97 Eci • CQ o E-i I 8 i •8 8 fd (0 (0 Q) iH i •• 8 < in I o r^ CM m oj ro O (Ti x> o *x> CT> 'I' 3" in fN M Xi Ti CT XI H j; XI jsi u ^< Xi ^ M M U M Si j:i M M U M Xi Xi Si M U OS M m Di fN H S < i < 2 • • • • • p p p o o o .H rH iH iH iH iH 03 CQ CQ CQ CQ CQ 1 jC 0) H (C •>H O 4J fe C Dl -- -H -H S 0) S CQ iH TJ 4J OQ Q) rH •a (1) rH (U rH 0) 4-) rH IH >i 2S8Z o .H c >i x: rg 0) 'tJ c (D en >,i X> m'ti in s 3 Q 03 pL4 .-( b ij M Qj in dj to S • E-i 04 CQ (d •O >i 0) 2 H ^ >,.H to W W V4 >
PAGE 114

98 C57BL/6 anti-BALB/c reaction The results presented in Table 21 compare the CML reactivity of primed B6 (H-2 ) cells generated against BALB/c (H-2 ) cells derived from either a mixed lymphocyte reaction or from animals undergoing GVHD from day 5. Both sets of GVH primed cells lysed those cells which express the host's K or D antigens as found on BALB/c, BIO.GD, BIO. A, B10.T(6R) and BIO.SOR), but did not lyse the B6 cells or unrelated third party strains such as B10.A(2R). lysed. The killing activity of GVH primed cells appeared to be identical, whether these cells resided in the spleen or the liver (Table 21 column 3 and 4). The surface phenotypes of both GVH populations are Lyt 1+ (60 to 70%), Lyt 2+ (20%) and Ig+ (20 to 30%). In addition, both sets of cells proliferate only towards the I-A antigen (data shown in Table 10). Thus, both sets of GVH primed cells apparently are identical, with the only difference being that they were isolated from two different organs. Prior treatment of the splenic cytotoxic cells with monoclonal anti-Thy 1.2 (HO-13-4) plus complement abolishes . 51 the ability of these cells to lyse Cr labeled BALB/c and BIO.GD targets (Table 21 column 5). Addition of anti-Lyt 2 antibody (53-6.7) but not anti-Lyt 1 antibody (53-13.3) to CML culture also inhibits the lysis of targets similar to

PAGE 115

99 8 ^ Ti 0) m G) C CM 0) l-l I 8 T y£> CQ •H O >i 4-1 •rH > •H 4J O (0 5 w s c 0) ^ jL) ro 2 a) c C ro rCM iH iH J3 fO XI 'O T! T3 XI J! T! X J>i tJ'J-i Si M Ti Xi M tTM J3 M ^ r3 M ty m XI ^ T) TD j^ cr m a; 05 vD cn < \8 < i^ w o o o o rH tH tH rH CQ pq CQ CQ 5i^ 0) xJ C (fl H W W >i W (0 OT 4J (U 8^ (C £ i I in a) H 05 CXrH •H > tn CO u c c: 0) -H (0 0) 0) (U VD X! rH ij CQ o &H a jJ CO 0) c • _, ij •rH 0) -a o m B e 4J ii 8gSK.>, ^3 O O (1) -H tJ y 03 (0 •rH (0 M U rH CS Eh 8.S to H G (!) 4-) gj -I fo S l-l IM (D 0) rH CO jj 4J a a) g Vj c H ^.rH >1-H C 2^8 iS g Q St:' CO •H x; -H u E-i Oj HH n> J3 i2 C £1 4J a 4-) (C Cn ai -H vj >, r-l CO C O 'Q Z3 d^ d) -rH rH O MH 4J •H w't> 0) ro U 4J 4J r^ E-i TS 3 -H C VO O 4J 4J •H m c to (C C T3 3 >14-) • >iO O W "^ CO 4J -H CO -H • CO (CJ 13 •H > en c a) 4-> -rH Q) gj Q) 4J (0 4-) 4J Qa >H 1C5 )-l O >i CO g) (1) • (0 U p 3 Vj "O 4J (D O W 4J 0) 0) M C CO >, 01 0) rH rH 0) (d IH 31 rH rH rH VH CO igg&88^S o

PAGE 116

100 that reported previously (102,103). In addition, cytotoxic cells derived from GVH animals must be maintained with exogenous interleukin 2; otherwise killing activity is lost, From these findings it appears that cytotoxic T cells are indeed present in either the spleens or the livers of animals undergoing acute GVHD across an entire H-2 haplotype. BIO.MBR anti-(A.TL x B10.MBR)F reaction In a set of experiments involving incompatibility at the H-2K plus D regions weak, but transient, CTLs developed in host organs undergoing severe GVH disease (Table 22). (A.TL X B10.MBR)F hosts were reconstituted with BIO.MBR cells. Cells from afflicted spleens of these mice were subsequently examined for CML reactivity from days 6 through 19. As can be seen in Table 22, weak but transient CML activity could be detected against target cells expressing the H-2D gene product, e.g., B10.D2 and BIO.TL during the early stages of the reaction (between days 5 to 8). However, by day 19 when severe GVH disease was obvious, no cytotoxic activity could be recovered from the affected spleens. In contrast, CTLs generated in MLR in the same combination exhibited strong CML reactivity against BIO.TL, BlO.S and s d B10.D2 target cells expressing H-2K and/or H-2D gene products.

PAGE 117

101 Table 22. CML reactivity of primed BIO .Mm anti-(A.TL x B10.MBR)F, cells generated either in iVlLR or in GVH. ferget H2 Genetics % MLR generated Specif (b) ic Rlease (a) GVH generated (c) Day 6 Day 22 cells K A J E D 40:1 20:1 40:1 70:1 BIO .Mm b k k k q -4 -3 -18 -12 A.TL s k k k d 26 18 12 1 BIO.S s s s s s 7 8 -10 -3 B10.D2 d d d d d 16 8 12 -5 BIO. BR k k k k k -1 -3 -12 B6 b b b b b 3 5 a) This test was performed in a 6 hr assay. b) Primary MLR consisted of BIO.MBR splenocytes incubated with irradiated (A.TL x B10.MBR)F, splenocytes for 5 days. The MLR prined cells were assayed at either 40:1 or 20:1 effector to target ratios. c) GVH reactivity was induced in sublethally irradiated (A.TL x B10.MBR)F, mice with BIO.MBR splenocytes. On days 6 and 22, the spleens were collected, nade into single cell suspensions and cultured overnight in interleukin 2 and assayed the next day. The day 6 GVH primed cells were assayed at a 40:1 effector to target ratio, v*iile the day 22 GVH primed cells were assayed at a 70:1 effector to target ratio.

PAGE 118

102 Histologically, the picture of GVHD in this combination was similar, although not as severe as in the entire H-2 difference. The number of recovered viable cells from the spleen was markedly lower than in the entire H-2 difference 5 with only 5 x 10 viable cells recovered per spleen as compared to 5 x 10 recovered in the entire H-2 GVH. Cytocentrif uge preparations of these recovered cells showed a marked predominance of PMNs (90 to 95%). From these findings it can not be ruled out that the cytotoxic cells have migrated out of the spleen and into the body and have contributed to the tissue destruction seen histologically. Ability of purified interleukin 2 to speed up mortality in K/D disparate GVHD In order to test the hypothesis whether increased cytotoxic activity could be induced in the K/D disparate GVH, two groups of (A.TL x B10.MBR)F, mice were reconstituted with BlO.MBR cells, with one group receiving 10 units of EL-4 derived IL 2 every second day up to day 7, while the other group did not receive any IL 2. The purified IL 2 had no mitogenic activity on resting splenocytes as determined by tritiated thymidine incorporation. In addition, this IL 2 had no deleterious effect on sublethally irradiated mice reconstituted with syngeneic splenocytes up to 30 days postengraf tment (Figure 20). The pattern of K/D

PAGE 119

CM -H H^S^ §S •H 0) C m 8 .ri u c g flj s cn.-i 0) C -H C

PAGE 120

104 o If) to N o in o c X > — c 4. O m C O (/) ^ O D < O O CJ Ixl < 01 z Ui (/) o Q. CO a ^o ^ o o in O "IVAIAdnS lN30H3d in

PAGE 121

105 disparate GVHD again exhibited the survival pattern as seen in Figure 5. However, those mice which received both the K/D allogeneic splenocytes and the IL 2 developed mortality rates which were comparable to an entire H-2 disparate GVHD. Thus, this exogenous IL 2 supplemented to a weak GVH significantly accelerates the disease process. The presence of CTL activity was examined in those mice developing acute GVH with the help of the exogenous IL 2. The results of such a study are seen in Table 23. This experiment reveals that cells obtained from the spleens from either the normal GVH or from the IL 2 supplemented GVH possessed identical equivalents of CTL activity. Similar results were seen in a combination involving only a D disparate GVHR, namely B10.M(17R) anti-A/J (Table 24). Histologically, the IL 2 supplemented GVHR had a disease which appeared to have progressed further than the normal K/D disparate GVH. Here the IL 2 supplemented GVH mice had increased amounts of liver necrosis along with increased numbers of infiltrating leukocytes in the parenchyma. BIO.AQR anti-(B10.T(6R) x B10.AQR)Fj^ reaction GVH induced by I region incompatibility did not result in readily detected cytotoxic responses. Figure 5 shows that such I region disparate mismatches resulted in the host animals dying with GVHD somewhat slower than hosts with

PAGE 122

106 'feble 23. CML reactivity of GVH primed BIO.MBR anti-(A.TL x B10.MER)F supplanented in vivo with or without interleukin 2. (a) "^ HK •2 Genetics A J E D % Specific Release (b) Targets Control 40:1 20:1 GVH 10:1 GVH + in 40:1 VIVO 20:1 IL 2 10:1 BIO.MBR b k k k q -18 -14 -9 -16 -17 -11 BIO.TL s k k k d 12 9 4 9 4 BIO.S s s s s s -10 -4 -5 -9 -10 -12 B10.D2 d d d d d 12 14 7 15 10 3 BIO b b b b b 5 -2 -5 -4 -4 -5 a) GVH reactivity v^as induced in sublethally irradiated (A TL x B10.MBR)F^ mice with BIO.MBR splenocytes. One half of the mice received 10 units of interleukin 2 i.v. on days 1,3 and 5. On day 6, the spleens were collected, irade into single cell suspensions and cultured overnight in interleukin 2 overnight. The cells were assayed at either: 40:1, 20: lor 10:1 effector to target ratios. b) This test was performed in a 6 hr assay.

PAGE 123

107 Table 24. CML reactivity of GVH primed B10.M(17R) anti-VJ cells supplanented m vivo with or without interleukin 2 (a) .— w^j.._l*jj.4j ^, \Cl/ I^get cells H-2 Genetics K A J E D % Specific Release (b) Control 17:1 GVH 8:1 GVH + in vivo IL2 17:1 8:1 B10.M (17R) A/J k k k d f k k k d d -7 20 -3 14 -4 -4 19 17 ^^ S?h''^fM^7P^'^^'"'^''''f '" sublethally irradiated VJ mice with B10.M(17R) splenocytes. One half of the mice received 10 TJilLf ^^terleukin 2 i.v. on days 1,3 and 5. On d^ ellhe c^^itZJ^''^-'^^^^'^^' "^^^ ^"^ ^i"5le cell suspensions and cultured xn mterleukm 2 overnight. The cells weS assayed at either 17:1 or 8:1 effector to target ratios, b) Tihis test v^s performed in a 6 hr assay.

PAGE 124

108 entire H-2 disparate GVH, but faster than the K/D disparate GVH. Yet, no cytotoxic activity against the host was detected in GVH on day 5 to 14. Representative data from day 6 is presented in Table 25. Again, as in the case of the K/D difference, the predominant cell type found in the spleen was the PMN (90 to 95%). To rule out the possibility that cytotoxic cells were present, but in low and undetectable numbers for functional tests, 650R irradiated (B10.T(6R) x B10.AQR)F^ mice were again reconstituted with BIO.AQR cells. These mice were sacrificed on day 6, their splenocytes removed and expanded for 2 weeks with IL 2. These cells were then tested in a CML assay and the results are shown in Table 26. The results show that CTL cells were present in these animals and these CTLs lysed B10.T(6R) B cell blasts (LPS stimulated) but not B10.T(6R) T cell blasts (Con A stimulated) (experiment 1). Besides a cytotoxic response directed towards cells expressing the I-A^ molecule found on B10.T(6R) B cells, cytotoxic responses were also demonstrated against several distinct target cells, e.g. BIO.AQR, BIO.A, B10.A(2R) and B10.A(4R). Since BIO.MBR cells were not lysed, the I-A^ molecule is probably not the target antigen. Further support that the antigen(s) seen by these GVH primed cells is not directed at the I-A*^ molecule was provided when antiI-A^ monoclonal antibody (10.2.16) did not block killing

PAGE 125

109 T^le 25. CML reactivity of primed BIO.AQR anti-(BlO.T(6R) x B10.AQR)F, cells generated in either MLR or in GVH. Specific Release (a) Target cells B10.AQR Con A blasts BIO.AQR LPS blasts B10.T(6R) Con A blasts B10.T(6R) LPS blasts In vitro (b) primed In vivo primed (c) -10 16 -11 -5 -11 -4 -28 -6 a) This test was performed in a 6 hr assay. b) The primary MLR consisted of BIO.AQR splenocytes incubated with irradiated (B10.T(6R) x B10.AQR)F, splenocytes for 5 days. The MLR primed cells were assayed at a 15:1 effector to target ratio. c) GVH reactivity vas induced in sublethally irradiated (B10.T(6R) X B10.AQR)F, mice with BIO.AQR splenocytes. On day 5 the spleens were collected, made into a single cell suspension and assayed the same day. The GVH primed cells v^re used at a 15:1 effector to target ratio.

PAGE 126

110 n t b X (0 VD <-{ d) P3 .-I ^ u I jk; j iH fl) CQ rS OS fO ^-4J E-" O • (U SB T3 (!) H O ^3 CD U m c (D M 0) n s iH m 55 .CM 4-) 0) c rH in -H (1) c u S3 (1) • -H U CO 3 (D T) •Q -P (1) C •H 1U-I to C C 0) fO s § 4J • rH §3 P C rH -H 8 >i •H >|4J ^^^O (U CO 4J 0! m c c (0 c c coo 1-8 -8 O M to U-l 1-1 ^ § § .g'g >1 ^J ^ •-' •H eg CO •H !^ fl) •H • f^ U O CO (T3 rH p 0) CQ fO M XI .H aC +J rH §•^8 (0 sii 3 -H -H 4J m m i-J
PAGE 127

in of the BIO.AQR LPS blasts by these CTLs (Table 27). Yet, this antibody is indeed capable of blocking PLT responses k directed against the I-A molecule using MLR primed B10.T(6R) anti-BlO.AQR cells. Further studies using these cytotoxic cells revealed that these cells were incapable of killing B6, BlO.F, BIO.M, B10.D2 and BlO.S LPS blasts (Table 28). These cells were tested on the 42 day after they were removed from the animals and showed that their lytic activity had changed. These cells now no longer had the capacity to kill BIO.AQR blasts as strongly as before but were now only able to kill BIO. BR blasts, both LPS and Con A blasts. It thus appears that this cell line originally contained at least 3 populations of cytotoxic cells: one directed against the B10.T(6R) LPS blasts, presumably I-A the second directed against BIO.AQR LPS blasts and the third population which was heteroclitic in nature which responded to BIO. BR LPS and Con A blasts. It should be pointed out that the addition of 20 units of purified IL 2 to the I region GVH reaction did not alter the rate of mortality in this system (Figure 21). These data presented in this combination demonstrate that a host versus graft (HVG) reaction is possible even though the host was irradiated with a near lethal dose of radiation (650R). The histopathological lesions of HVG are identical to those

PAGE 128

o nj ii Eh G >1 •H iS S "5 P C iH m -H cnoJ fl) 0) (0 fl) C 0) O ft) en S )H S C T> r-tO O C CQ o aj o -H CN VD ^ IW fO w O d Eh x: 0) _, q -H 4J g •H -ro 3 -H 4J O £ ^1 iHS CN JJ -^ tH T) > UJ Ql M > fQ O fiC C -H • O (U -l 03 in S O d) ^ ^ .,H Q 2.-'H x: §"^li

PAGE 129

113 o in O I > m CSJ o Z X o > u o o CO Z UJ u. < O z Ul cn o < .1 O o o o O CO o o ID O in o o ro O CO ^ IVAlAdnS lN30H3d

PAGE 130

m Table 27. The inability of monoclonal antibodies to block killing of BIO.AQR LPS blasts by GVH primed BIO.AQR anti-(BlO.T(6R) x B10.AQR)F^ cells expanded in vitro with interleukin 2. (a) % Specific condition Release (b) % Inhibition no addition 57 — 10.2.16 1:20 (c) 10.2.16 1:40 65 59 B 312 1:20 (d) B 312 1:40 52 54 9 5 14-4-4 1:20 (e) 14-4-4 1:40 64 59 anti-Lyt 1 1:10 anti-Lyt 2 1:10 anti-lhy 1:10 71 50 71 9 a) GVH reactivity vas induced in sublethally irradiated (B10.T(6R) X B10.AQR)F, mice with BIO.AQR splenocytes. On day 6, the spleens were collected, made into a single cell suspension and cultured in interleukin 2 every second day. The experiment was done on day 29. The effector to target ratio was 10:1. b) This tes^ v^as performed in a 4 hr assay. c) Anti-I-A antibody (10.2.16) with a final dilution of 1:20 was used to block lysis. d) Anti-K antibody (B 312) with a final dilution of 1:20 was used to block lysis. e) Anti-I-ET antibody (14-4-4) with a final dilution ofd 1:20 vas used to block lysis.

PAGE 131

115 to I g OQ X **-*. ^--* m a; yo %,^ E- CM o C r-l •f-t m -^ 1 a) • -H 1-1 00 4J i-i '"^ B Q) c 1-1 a; •H S^ jr H • 4J o •^ m T) 'CJ (1) i -H •-H (rt >-l n, U4 X •8 •H 1i iH u •H U OP -O CM C iH e TO VD W (0 o -H 9i CN O iH 4J C iH •H CM rH Q 0) CO tJliH ^8 CM iH I m 1^ I I I I I r^ o I (-^ rI I I I I r^ iH I CM W3 I I I I I I CN fN I in I I '3' c rn v£) •-• I I I I I cr. I en c o o r-l iH CQ OQ CT) I I H CT> CN CM T3 M M CO ja D414-1 T3 to o C) o 0000 mCQ CQOQCQcQcqCQ 1 i rH +j 8 c •H c •H >1 -s s JS 4J UM^ (1) •d c 6 -H to rH fa nj S o O r-l (U 2 73 ^ 4J Eh O • 0) (1) V-i •H O •73 to O (0 C (D M gj to M "-I > • m -H rH (1) 0) >i m M-l ^ iJ CM to >, >, 4-) to 17 X? 0) C (0 M UJ iH in -H O -H ffi fd (U fO iH <^ C M •H C 0) (0 O 4J c c 'O c c o o 0) • -H -H tj J-' (1) ^ O TJ d to CO S iH fl) a c Q44J +j t-1 -H to m ni •H < m ML) u • D<^ m m o o to to a ft, S CQ W -D 3 3 JS iH CO CO CO > -H m 2 2 id O 3 O Eh Eh Eh 'S'S

PAGE 132

. 116 produced in GVH (104); thus, it could be argued that in this combination both types of reactions are occurring and that the CTL activity is undetectable because the Cr assay does not detect the activity of these few cells. It is interesting to note that the rate of mortality in B10.T(6R) anti-(B10.T(6R) x BlO.AQR)Fj^ combination is identical to that BlO.AQR anti(BIO .T( 6R) x B10.AQR)F^ reaction (Figure 22). Other I region mismatches If A second combination involving an I-A mismatch (BIO X BlO.Q}Fj^ anti-BlO.MBR has also been examined. As can be seen in Figure 23 the mortality rate in this combination proved comparable to an entire H-2 mismatch [(BIO x B10.Q)Fj^ anti-BlO.BR] When the spleen cells from these I region GVH mice were grown in vitro in the presence of IL 2 for 1 week, CTL activity was demonstrable against both the donor haplotype cells B6 and BIO.Q, as well as the host cells BIO.MBR (Table 29). Another I region mismatch namely (BIO.MBR x B10.GD)F^ reacting against BIO, an I-A disparate reaction has also been studied. Figure 24 illustrates the mortality kinetics of such a reaction. (B10.A(4R) x B10.GD)F, anti-BlO provides both K and I-A differences and demonstrates a strong lethal GVH reaction. In contrast, the I-A

PAGE 133

(0 X t! to U --; (3 I— I fO V£> W U "T! • H S ;^ O 0) • w 05 3 -r-l o vo t3 r^ .H O -^ C CQ in Eh •r^ (D — vo w o w p TJ x: ^;^ r-i n (U 4J CQ rH S (0 "^ • ^ "fQ t3 -H oi 5 •iH -H rH O "O n3 >i (0 (0 rH U Q) iH rH i-H nj J3 (0 -H S5 ^ (1) w C r-l (U H XI 4J CQ c
PAGE 134

118 O lO lO ^ II II c c O UJ JL I— > M O LlI IZi W z o o liJ a: lo d m '^< lo cr o < o 00 O ro hz LlI :s H Li. < CD iij H CO o CL P CO O o o IT) o in CVi nVAIAdnS lN30d3d

PAGE 135

o CQ H >i C tH 4-) O J 4-1 s • fa j2 >H -H jj m TJ W CD O 0) 8 4-) m O c 3 C 4-1 -H OJ c B 0) O 14-1 c "'d S^ o >i C .H 4-) "3 O -H m ^^5 8 8 H .j-( 4J 0) > ni U U V4 O < -H -H U) i^ y .1. •-' ^ W Ci^ trt P -H 54 4-5 ^ i^ § ^ 8 c > W -H O O 0) G -H 4J 4J JS O • •-( 4-) C O ^ M .H CQ D CO (DO) rM r-lU 4J •rH rH >^ .r4 py PQ -H fd

PAGE 136

120 m a d 2 S o to a in o IL* Q o a o X a O X a S H a a III III o E r bl u U. U. o -< Q < Ul t< a: o z u I(/) o CL 10 o o o in in nvAiAyns iN3oy3d

PAGE 137

121 T^ble 29. CML reactivity of GVH primed (BIO x B10.Q)F^ anti-BlO.MER cells expanded with interleukin 2. (a) Target cells H-2 Genetics K A J E D B6 LPS blasts b b b b b BIO.Q LPS blasts q q q q q BIO.MBR LPS blasts b k k k q BIO.MER Con A blasts BIO.SOR) LPS blasts s s s s s % Specific Release (b) 14:1 (c) 13 18 22 17 -4 a) BIO.MBR mice were sublethally irradiated and reconstituted with (BIO X B10.Q)F, splenocytes. On day 5 these mice were sacrificed and their spleens were ranoved. A single cell suspension of cells vras then fed purified interleukin 2 every second day for 1 week. b) This test was performed in a 6 hr assay. c) The effector to target ratio was 14:1.

PAGE 138

U JJ — 03 •H Q ei s 13 -u o g 0) m • s >i 4J c o • *J rC O rH p -H •H U fQ --I iH ^0 p „ IB (d (0 S4 X >JJ >1 CO +J "sr rH +J (1) J-) O -U Vj (1) C O a. .-I (u 0)

PAGE 139

123 O in o o O o O X o o t m m $ X X n ^f^ lU on o: 1^ CO 13 < ^< s o 1— • O-Q 1n — < if> CD ^ OQ n z ^-' ^-^ o o • LU o: 1^ II c O ro O CO V ^ :e Ll. < cr CD LU Iin o CL CO >< o o o lO ID IVAIAdnS lN33d3d

PAGE 140

124 differences, (BlO.MBR x B10.GD)F^ anti-B6, only showed a mild reaction. When the splenocytes were recovered and expanded in vitro for 1 week with purified IL 2 a weak cytotoxic response was directed against the B6 LPS targets, while little if any lysis was seen against either of the donor cell lines (Table 30). It would appear that in this reaction only a graft versus host reaction was occurring and that the weak mortality which was demonstrated in vivo is the result of a weak cytolytic response of T killer cells. In contrast, in GVH reactions, such as BlO.HTT anti(BIO.TL X B10.HTT)F no mortality was observed (Figure 25). When GVH primed cells from these animals were grown in the presence of IL 2 and tested for CTL activity, no detectable lympholysis could be observed (Table 31). Minor histocompatibility antigens The presence of in vivo CTLs was next explored in those mice which produced low but detectable PLT cells towards minor histocompatibility antigens. Minor histocompatibility disparate GVH reactions were established using the DBA/2 mouse as the host. Survival times of DBA/2 mice reconstituted with BALB/c, B10.D2 and BALB/c are presented in Figure 26. DBA/2 hosts grafted with BALB/c splenocytes showed only 8% mortality over the first 45 days, while the DBA/2 mice receiving the B10.D2 cells showed a

PAGE 141

125 'feble 30. OVIL reactivity of GVH primed (BlO.MBR x B10.GD)F, anti-B6 cells expanded with interleukin 2. (a). % Specific Release (b) Target cells 16:1 8:1 4:1 B6 Con A blasts 4 1 1 B6 LPS blasts 9 6 5 B10.MBR Con A blasts -1 -3 -1 B10.MER LPS blasts 2 4 BIO.GD Con A blasts 3 2 4 BIO.OD LPS blasts 1 a) GVH reactivity vas induced in sublethally irradiated BlO.MBR mice with (BIO x B10.Q)F, splenocytes. On day 5, the spleens were collected, made into a single cell suspension and cultured in interleukin 2 every second day. The cells were tested on day 14. b) This test was performed in a 6 hr assay. The effector to target ratios used were:16:l,8:l and 4:1.

PAGE 142

< M ^-^ iH • ffi T3 -^ (U $ Eh -H O 4J M Tj rH (C ffi OQ vj • 0) C (0 1 (0 O >i 4J j:: rH _23 u 4J m 'u O 0) -^ •I— I -r5-i iiiS

PAGE 143

O) O) 127 o in X < O X I g CO o OQ X < X -J o m g £ lo lo 1X K X o ID g m • O o O CVJ Ul < o z Ul ho Q. CO O o SP o nVAIAdnS iN33d3d in

PAGE 144

Table 31. Lack of CML reactivity of GVH primed BlO.HTT anti-(B10.TL x BIO.HIDF, cells expanded with interleukin 2. (a). 128 Target H2 Goietics % Specif LC Release (b) cells K A J E D 40:1 20:1 10:1 BlO.HTT Con A s s k k d -11 -5 -9 BlO.HTT LPS -9 -10 -4 BIO.TL Con A s k k k d -7 -5 -3 BIO.TL LPS -14 -12 -12 BIO.SOR) Con A s s k k d -4 -6 -5 BIO.SOR) LPS -1 -3 -3 DBAA LPS q q q q q -9 -3 -3 SFA LPS d d d d d -4 -6 -4 a) GVH reactivity vas induced in sublethally irradiated (BIO.TL x B10.HTT)F^ mice with BlO.HTT splenocytes. On day 5, the spleens vre collected, made into a single cell suspension and fed interleukin 2 every second day for 1 week. b) This test was performed in a 6 hr assay.

PAGE 145

0) p T) -H 4-) • (0 W ; a to 0) u s^ -H CN iH (C —I rH "^ O "C

PAGE 146

130 =U tf) M U J| UJ U O CO O ~^ ^Q O 03 OQ CD Z O • -4 IO a. o CM OT o o o o o o o o o o o 0) CD t^ (0 m t K) CM nvAiAans iN33a3d

PAGE 147

131 mortality rate of 37% by day 13. However, even in this latter set some mice survived several months. In contrast, a different picture was observed when the BALB/c mutant mouse strain, BALB/c was used as the source of the donor cells. The BALB/c mouse does not express the H-2L molecule, so any reactivity which occurs is directed against both the minor histocompatibility antigens of DBA/2 and the H-2L molecule. This reaction resulted in a complete mortality by day 34 with the majority of the mice dying by day 9. When CTL activity was assayed in the BALB/c anti-DBA/2 combination on day 5 of the GVHR and after culturing the cells in IL 2, no significant killing was observed (Table 32). If the GVH primed cells were tested in CML the same day they were removed from the host animals so that the contaminating PMNs (which comprised 85% of the total viable cells) were present no killing was observed even in a 48 hr assay (Table 33). Thus, PMNs and non-MHC primed cells were not capable of killing ^^Cr labeled targets. The survival curve of sublethally irradiated AKR mice reconstituted with CBA/CaH cells is presented in Figure 27. Three out of 5 mice died within the first 17 days. Those animals not showing acute GVHD survived several months without any indication of disease. This survival was similar to that seen previously in the K/D disparate GVHD.

PAGE 148

132 teble 32. CML reactivity of primed BALB/c anti-DBA/2 cells generated in GVH •terget cells H-2 Mis % Specif ic Release (a) 20:1 BALB/c Con A d b 8 BALB/c LPS d b -2 DBA/2 Con A d a -5 DBA/2 LPS d a -1 a) This test vas perfomned in a 6 hr assay. GVH reactivity vas induced in sublethally irradiated DBA/2 mice with BALB/c splenocytes. On day 6, the spleens were collected, made into a single cell suspension and cultured overnight in interleukin 2 prior to use. The GVH primed cells were assayed at a 20:1 effector to target ratio.

PAGE 149

133 T^le 33. CML reactivity of BALB/c anti -DBA/2 cells generated in GVH Target CPUs H-2 Mis % Specific ] 24 hrs Release (a) 48 hrs BALB/c d b DBA/2 d a -1 B10.D2 d b -1 B10.BUA16 w b -1 -3 a) This test was performed in either a 24 or 48 hr assay. GVH reactivity vas induced in sublethally irradiated DBA/2 mice with BALB/c splenocytes. On day 6, the spleens were collected, made into a single cell suspension and used the same day as they were removed from the animal. The GVH primed cells contained 15% lyni^ocytes and 85% EMNs. These cells were then used at a 50:1 effector to target ratio.

PAGE 150

m 3 IS 0) CD _2J O 4J 'O *^^ a ^ b

PAGE 151

135 o o o lO z UJ Iu. < ir o o a. o CO o o o o in ir> CM "IVAIAdnS iN30y3d

PAGE 152

136 Histopathological examination of the AKR mice suffering from acute GVHD revealed an identical disease picture as described earlier (Section 3.2 F). In addition, when the spleen cells of these GVHD affected mice were tested on day 10 for cytolytic activity, CML was exhibited on AKR targets, with weaker activity on DBA/1 and DBA/2 target cells (Table 34). Although the CML activity appears to be directed against Mis this may not be the target since CBA/J, which is now thought to possess the Mls^ determinants, was not killed. In addition, no CTL activity against Mls^ was observed in the BALB/c anti-DBA/2 system. Thus it is likely that the CBA/CaH cells are detecting another antigen which IS present on AKR cells. KAbility of primed lymphocytes to cause mortality in sublethally irradiated mice In an attempt to determine whether GVH primed lymphocytes could be passaged into secondary hosts, it was discovered that lymphocytes primed either in in vitro or in vivo have the ability to cause mortality in both syngeneic and allogeneic sublethally irradiated animals. This is presented in Figure 28 for B10.A(5R), BlO.GD and BIO.S mice which were sublethally irradiated and injected i.v. with 1 X 10 (B10.A(4R) x B10.GD)F, anti-BlO primed cells generated in vivo As can be seen, these

PAGE 153

137 Table 34. CML reactivity of primed CBA/Ca anti-AKR cells generated in GVH Target cells H-2 Mis Specific Release 40:1 (a) 20:1 AKR k a 33 39 DBAA q a 16 12 DBA/2 d a 13 10 CRA/J k d 7 8 BIO. BR k b 5 CRA/Ca k b 11 a) This test was performed in a 6 hr assay. GVH reactivity was induced in sublethally irradiated AKR mice with CBA/Ca splenocytes. On day 10, the spleens were collected, made into single cell suspensions and cultured overnight in interleukin 2 prior to use. The GVH primed cells were assayed at either 40:1 or 20:1 effector to target ratios.

PAGE 154

s ^ c w o • X o Ti d) r-i .^* 0) $ CQ ^ Vj s b^ &j Be • s CQ CO to o v^ iH iH >1 r-\ COD 4.) 8 o •H ^ t-t i-l T! S x: 5 in < U • ChO T3 • i-H -; ^ k 0) < fd • -H O CD O TJ in rH H (0 VO n^ £bx:8 O >i 3 S iH r^ s • iH TJ -H 73 >y <0 id OJ O j= 4J a--H 4J (C 'O >i to CQ m •H 55 >^ -rH Q ^ -H JJ S ^ -H >. le J 00 >ir-l i-tU CM jJ HJ Dy •H jC — >i >-i m 0) 8 -S •H Q ^ rH x: fo S CO CQ H

PAGE 155

139 :3 t fe H < < < uj 5 o qQciO _o £5 UJ _ S < L.<|-< t< o 222922 X CD CD (Z3 03 (S (Q ^ > O* O X" o o in O in CM o z to y < C3 to O (f) O nVAIAMnS iN30a3d

PAGE 156

140 anti-K I-A cells proved lethal for both the allogeneic mice B10.A(5R) and the syngeneic mice BIO.GD, with death occurring between days 10 and 17. The primed cells could even be irradiated with 2000 R of irradiation and yet still cause mortality in these animals. Third party mice, such as BlO.S, also died between days 15 and 22; however irradiation of the primed cells increased survival time of the BlO.S mice even though they eventually died by day 32. It should be noted that the primed cells had no effect when injected into nonirradiated mice, either syngeneic or allogeneic. In a follow up study using BIO.RIII anti-BlO.M primed cells generated in vitro similar results also occurred. Both syngeneic and allogeneic mice died between days 10 to 20 after engraftment. Supernates of the primed cells also caused mortality in sublethally irradiated mice. Interestingly, if primed cells were cultured overnight in fresh medium, this supernate did not have the ability to cause mortality in sublethally irradiated mice. In addition, normal unirradiated mice were not susceptible to mortality induced by either the primed cells or culture supernate even up to 3 months after injection. Neither the supernate nor the cells could induce pathological effects to an embryonic mouse fibroblast culture, indicating that such factors as cytopathic viruses, were not playing a role in mortality.

PAGE 157

1^;, 141 L. Histology and pathology of secondary disease Mice receiving primed cells from the previous experiment were examined histopathologically on day 10. When BIO.RIII anti-BlO.M primed cells were injected into sublethally irradiated BIO.M mice, the normal histology of acute GVHD was observed in the liver. Perivascular cuffing and leukocytic infiltrates associated with the veins and bile ducts were present together with the parenchyma damage (Figure 29). In the lungs of these animals some edema and mild hypercellularity was also observed (Figures 30 and 31), In marked contrast, when BIO.RIII anti-BlO.M cells were injected in irradiated syngeneic hosts, a different pathology was observed. The lungs were markedly more hypercellular with much more edema and congestion (Figure 32). In addition, the spleen also showed marked hypercellularity (Figure 33). Presumably, large numbers of the injected cells became trapped in these two organs. In the liver, leukocytic infiltrates were found infrequently. Such infiltrates, when found, appeared in the parenchyma per se and not along the central vein or bile ducts (Figures 34 and 35). This appears to be similar to a hepatitis type lesion, and not like the pathology observed in the livers of hosts with allogeneic GVH disease.

PAGE 158

Figure 29. The liver of a BlO.M aninal undergoing GVHD fran printed lyn^ocytes: BIO.RIII anti-BlO.M. This liver shows perivascular cuffing of invading leukocytes. Tte leukocytes are radiating frcm a central vein and appear to be invading the parencyita. Figure 30. The lung of the BlO.M animal previously examined. This lung ^ows little abnormality in it. A little ectema is seen but this is similar to that seen in the irradiated control

PAGE 159

143

PAGE 160

Figure 34. The liver of a sublethally irradiated BIO.RIII mouse reconstituted with prined BIO.RIII anti-BlO.M lyitphocytes. Leukocytic infiltrates are ooramonly found in these treated animals. No perivascular cuffing is observed. The parenchyma of the liver appears slightly edematous. However, unlike the lesions of GVHD there does not appear to be any massive amounts of necrosis. Figure 35. A higher magnification of the previous liver. Tte farenchyma appears edematous, but is essentially intact. Tte hepatocytes show \ell ^fined itembranes. A leukocytic infiltrate is daserved at the upper left corner, but this is a different type of infiltrate as compared to figure 10. Necrosis is not readily apparent.

PAGE 161

145

PAGE 162

Figure 33. The ^leen of the aninal previously examined. The spleen is hyerchronatic due to the hypercellularity, massive amounts of mononuclear cells are present. Little normal architecture is seen due to the crowcted conditions. The pathological change is much different fron that seen in Figure 15.

PAGE 163

147

PAGE 164

Figure 31. Ths lung of a normal irradiated mouse 14 days after irradiated. The alveolar walls appear thin with little edema. The alveolar sacs are evenly spaced. Very few mononuclear cells are found in the alveoli. No gross morphological changes are seen. Figure 32. Tte lung of a BlO.RIII mouse reconstituted with priited BlO.RIII anti-BlO.M cells day 14. Tte alveolar walls are substantially thickened and congested. The tissue is edematous, but there is no cellular exudate present. Monocytic cellular infiltrates can be discerned in the alveolar walls. This process is indicative of interstitial jxieumnitis.

PAGE 165

149 -J '.*• '.ff St ^< ife 1/7 ^ i, Iglj ^;

PAGE 166

150 M. Mortality induced by anti-I-A^ long term cultured T cell lines and cloned T cell lines The long term anti-I-A^ cell lines (BIO.RIII x B10.AQR)F^ anti-B10.T(6R) and (BIO.S x B10.AQR)F, antiB10.T(6R) have been maintained for over two years by Dr. A.B. Peck through sequential restimulations (presently 42 restimulations) with B10.T(6R) cells. At the time of the 25 restimulation, the (BIO.RIII x B10.AQR)F, antiB10.T(6R) line was cloned using limiting dilution procedures. Of fifty growth positive microtiter plates with "presumed" clones which were transferred to culture flasks for expansion, ten grew well enough for study. Analyses of the reactive patterns of these cells are presented elsewhere (105). The parental lines and the cloned lines have been maintained without the use of exogenous IL 2. The cells are Lyt 1+2, react strongly in secondary MLRs, are noncytotoxic and produce T helper lymphokines. Reconstitution of (B10.T(6R) x B10.AQR)F, with 10^ parental or cloned T cells has been found to induce severe GVH like disease resulting in death of the host between days 10 to 18 (Figure 36). One cloned line, 4, however, proved highly inefficient in inducing lethal GVH disease. This cell line is known to react primarily against the I-A^ molecule cross reacting against the I-A^ gene product (105). Each host animal displayed wasting disease, diarrhea and

PAGE 167

X d) +J O C B8 (0 8S^ 8 'S-i 1— iC (C -r-l w ^ a) £ <; c (jj jq i-3 rH --CQ£> O Oi +J rH -H CQ -3 X • T) — ~ O 0) pi 03 '-• iH 4-> >^ P3 3 w o VO ^ 4.) &^ iH O •H ^43 0) — c o •H in •-4 '0 3 Eh • 4J o m • w ra tH .. m c CO '^ = 01 ^ O -H n c Q) 0) -HC £ tu -H O a) >a en 0) e; 13 .. ^ •H I— I O O >^r-^ tJ 03 U iH x> •H a 13 4-) -^ ^ 0) C Eh >:< r-ti ?^ • JJ E (C M O •H iH -H fll iH <-^ diTi CucQ (0 43 (0 I 4-5 >H -"-H y (n M CO 4J S C -H CK c • iH 0) rX>OV iH 0) fcj ro 1 ^^ T3 i| a) 4-) 4J (C c (0 "O M 0) Vh )-i •H O >1 ^ rH H to 4J Q) >i Xi -rH 5 d) 'O f*l Id a 35 oa u

PAGE 168

152 O (J) ro 03 ro ^ ~ rO ^ in ro \ V in CO H ^ Z 'd' — w UJ 2 ro 1CM < — CVJ CD z UJ — c\j 1o O a. CJ m — 2] q: 00 Q. X a: g NH a. < UJ o fO ID Q U_ If) — in o UJ m ^ tro ^ -o in w rO lO O in — O O in cn CD lO O 1 1 1 ro 1 3Nn a3NonD ao "iviN3yvd

PAGE 169

153 hypothermia. Histologically, the liver, lungs and intestines revealed the pathology of GVH disease. Somewhat unexpectedly, however, these same cloned lines have been found to be capable of inducing disease in sublethally irradiated BIO.AQR hosts. However, histopathological examination has revealed a greater involvement of the lungs, with concommitant less involvement of the liver and intestine. The common lesions of GVHD are not present in these animals. These latter points suggest the disease process may be different and has nonspecific components. Not every long term primed cell line is capable of causing mortality in sublethally irradiated mice. For example, the B10.D2 anti-BlO line produced by Dr. B. Elliott and maintained by Dr. A. Kimura, when injected into B6 mice no mortality was observed a 60 day observation period (data not shown). This cell line does not produce large amounts of detectable soluble lymphokines (Peck and Kimura, personal communication); thus, the lack of mortality may be due to the fact that lymphokine production is limited and as a result no lymphokine mediated responses are elicited. However, this hypothesis remains at best speculative.

PAGE 170

154 3.3 Attempts to Modify GVHD A. Attempts to prevent lethal GVHD using concentrated monoclonal antibodies directed towards the host 's I-A molecule ~ Results obtained from the GVH and PLT data suggested that the I-A molecule is crucial in the development of lethal GVHD. It was decided, therefore, to determine if blocking the recognition of the I-A molecule of the host by using concentrated monoclonal antibody would modify subsequent GVH. The system examined was as follows. One milliliter of the purified monoclonal antibody (10.2.16 generoulsy provided by Dr. E. Wakeland) at a concentration of 1 mg/ml was injected i.v. into sublethally irradiated BIO. A mice 2 hours prior to receiving the B10.A(5R) donor cells. Monoclonal antibody 10.2.16 binds the I-A*^ molecule expressed in BIO. A mice. Survival was then used to indicate whether the antibody had any beneficial effects. The results, shown in Figure 37, indicate that both the control GVH animals and the experimental animals succumbed at the same rate. Histological studies revealed that both sets of mice had similar lesions. Under the present test conditions, it would appear that GVHD was unable to be controlled by this protocol. This combination differed genetically at both k k I-A and K This K difference, while insuring a lethal GVHD, may have distorted the final interpretation.

PAGE 171

U Q m 8 ^8^ CM 8 0) •d >il s H fc, &^

PAGE 172

156 OT -r d V> £ 3 -J t O -1 % X uj K> o H Q t^lUl -. (K o g: o < X V> OD > ID TRAN ANTI BIO -I O UJ fr jc K K < < O ro Z o o d Ul 5IOCD c -lO
PAGE 173

157 B. Attempts to prevent lethal GVHD using neonatal splenocytes i. Inhibition of acute lethal GVH disease using CBA/J newborn spleen cells The experimental protocol utilized in this study to test for the ability of newborn spleen cells to suppress lethal GVH disease in irradiated host mice reconstituted with allogeneic adult cells was as follows. Donor adult spleen cells were incubated 24 hours in the presence or absence of newborn spleen cell populations prior to engraftment in sublethally irradiated host mice. Newborn spleen cell populations were suspensions of whole spleen cells which had incubated in culture plates 12 hrs. After 18 hrs of coculturing, 30 x 10 viable cells were injected i.v. into host animals. Control reactions included donor adult cells cultured in medium alone or cocultured with adult spleen cells syngeneic with the newborn cells. Results from two sets of experiments in which CBA/J newborn cells were cultured with adult BIO. BR cells prior to engraftment of semi -allogeneic (BIO x B10.BR)F, or allogeneic B6 hosts are presented in Tables 35 and 36, respectfully. Host animals receiving adult BIO. BR cells or BIO. BR cells cocultured with adult CBA/J spleen cells died of lethal GVH disease between days 6 and 12. In contrast, only 33% of the host mice receiving BIO. BR cells cocultured

PAGE 174

158 with CBA/J newborn spleen cells at a ratio of 10:1 died within 12 days, while 66% of the mice showed long term survival. Similar survival rates were observed for mice receiving BIO. BR cells cocultured with CBA/J newborn spleen cells at ratios of 14:1 and 28:1. Data presented in Tables 35 and 36 reveal a number of additional points. First, newborn spleen cells alone were unable to induce lethal GVHD. Second, the suppression of GVH by the newborn spleen cells could be diluted out at adult to newborn cell ratios greater than 56:1. Third, CBA/J newborn thymus cells were unable to suppress the GVH reactivity of adult cells. Fourth, whereas AKR newborn spleen cells were as effective as CBA/J cells in suppressing subsequent GVH in B6 hosts, (BIO. BR x SWR)F^ newborn spleen cells proved inactive in (BIO x B10.BR)F, hosts. ii. Histopathology of the experimental and control host animals All mice which died were examined histologically to verify the presence of GVH disease. In addition, mice showing long term survival were occassionally sacrificed at various times and examined histologically for signs of GVH disease. In control and experimental animals which died or were presumed to die prior to day 14 to 15 postengraf tment classical symptoms of GVHD were noticeable in histological

PAGE 175

159 Table 35. Suppression of lethal GVHD by rewbom spleen oells in (BIO X B10.m)F, host animals reconstituted with semi -allogeneic adult Bid. BR spleen cells. Responding cell population Survival Tines (a) BIO. BR 7, 8, 9, 10, 10, 10, 10, 10 [9] BIO. BR + CBA/J Adult cells 7, 9, 12 [9] BIO. BR + CBA/J Newborn spleen cells 8, 12, 50+, 50+, 50-f, 50+ [45+] Newborn CBA/J spleen cslls 50+, 50+, 50+ [50+] BIO. BR + Newborn (BIO. BR x SWR)F, spleen cells 7, 7, 7, 8, 8, 9, 9, 9, 9, 10, 10, 10, 10, 11 [9] a) The number represents the day postengraftmsnt that a given animal died. Underlined dates represent mice that have lived considerably longer than control GVH animals. The number within [ ] indicates the mean survival time of a given series of animals. The symbol + indicates that those animals were still living after that day and did not appear to suffer from GVHD.

PAGE 176

.':?^y160 Table 36. Sippression of lethal GVHD by newborn spleen cells in B6 host aninals reconstituted with allogeneic adult BIO.ER spleen cells. Responding cell peculation Survival Tines (a) BIO-BR 6, 7, 8, 8, 10, 12 [9] BIO. BR + CBA/J Adult cells 7, 8, 11 [9] BIO. BR + CBA/J Newborn thymus cslls 5, 6, 6, 7 [6] BIO. BR + CBA/J Newborn spleen cells 10:1 (b) 10, 14, 25+, 25+, 72, 74 [36+] 14:1 9, 30+, 30+, 40+ [27+] 28:1 8, 30+, 30+, 40+ [27+] 56:1 10, 11, 30+, 30+ [20+] 100:1 7, 9, 10, 10, 12 [10] BIO. BR + CBA/J Newborn spleen supemate 6, 7, 16 [10] BIO. BR + AKR Newborn spleen cells 8, 20+, 20+, 20+ [17+] a) The number represents the day postengraftmsnt that a given nouse died. Underlined dates represent mice that have lived considerably longer than control GVH animals. The number within [ ] indicates the nean survival tine of a given series of animals. The symbol + indicates that those animals were still living after that day and did not appear to suffer fran GVHD. The ratio of adult :newbDrn oslls used in a given experinent. b)

PAGE 177

161 sections beginning on day 5 to 6 Such animals exhibited '~ wasting, diarrhea, a hunched posture and hypothermia. Of the long term surviving animals, approximately 30% developed symptoms of chronic GVHD. Between days 35 and 50 postengraftment, these mice became lethargic, developed skin lesions, and showed some fur loss with associated redness of the skin (Figure 38). Histologic examination revealed leukocytic infiltration in the dermis of the affected areas. In addition, the liver often showed abnormal histology, including a "punched out" appearance similar to chronic hepatitis, loss of hepatocytes, and disruption of the liver chord pattern (Figures 39 and 40). No gross abnormalities were seen in either the spleen, kidney or intestines. iii. Functional reactivity of donor cell populations after initial period of culturing To control for the possibility that the culturing of adult cells with CBA/J newborn spleen cells led to a functional inactivation, these mixed donor cell populations were tested for their mitogenic responsiveness to the T cell mitogen Con A and the B cell mitogen LPS. As shown in Table 37, adult BIO. BR cells cultured alone, as well as BIO. BR cells cultured with newborn CBA/J cells responded strongly

PAGE 178

Figure 38. The skin of a CBA/J nevtorn suppressed mouse day 72. This skin belonged to the mouse examined in the previous micrograph. By day 25 this mouse began to lose soitB of its fur and v^ere it did lose it, the skin was red. In this figure leukocytic infiltrates are found in the dermis (lovrer left corner), vhile the epidermis appears normal and acellular. From this finding it appears chronic GVHD exists in the animal.

PAGE 179

163 :^ '. -?-, km.\^

PAGE 180

Figure 39. The liver of a CBA/J newborn G7H suppressed nouse day 72. This (BIO x B10.BR)F, aninal was reconstituted with BIO. BR splenocytes treated with CBA/J nevfeorn suppressor cells. This animal survived much longer than did the GVH control mice v*io died by day 14. This liver demonstrates a "punched out" lesion. Here nuiterous vacoules are seen within the vacoule is a nucleus. Figure 40. A higher magnification of the previous micrograph. This micrograph reveals the heterogeneity of the nuclear ramnts. Tte liver is obviously different from the acute GVHD liver pathology seen in Figure 10.

PAGE 181

165

PAGE 182

Table 37. The ability of BIO. BR splenocytes incubated with CBA/J newborn splenocytes to respond to mitogens. 166 H-TdR Incorporation CPM+SD (a) cell population no addition BlO.m untreated (d) 5977+ 899 9559+ 908 BIO. BR cultured alone for 18 hr (e) BIO. BR with CBA/J NB 13067+1642 cocultured 18 hrs (f) Newborn splenocytes cultured alons 18hrs (g) 2059+461 Con A (b) 2314+ 609 LPS (c) 45398+13849 64360+6330 96703+8313 93245+3892 94637+1540 90382+3493 2302+ 286 ^^ f JcfS ^!?d?^^"^^^ ^^ '^ ^" folloving a 12 hr pulse with b) Concanavalin A dose was 5 Mg/ml. c) Lipopolysaccharide dose was 100 pg/ml. S?;^ splenocytes v^re removed fron the animal iimediately e) 10 BIO. BR splenocytes <^re incubated for 18 hrs in untreated nedia. f ) 10 BIO. BR splenocytes ;,ere incubated with 30 x 10^ CBA/J newborngsplenocytes for 18 hrs in itedia. ^dia "^"^^^^ splenocytes ^^re incubated for 18 hrs in

PAGE 183

167 toward both Con A and LPS. Thus, the donor cell populations were still functional at the time of injection into the host mice. iv. Functional reactivity of lymphocytes derived from long term surviving host mice As demonstrated earlier, protocols have been described which permit functional testing of cells derived from organs undergoing severe lethal GVH reactivity. Using these procedures, it was possible to examine the proliferative and cytolytic reactivities of cells residing in the organs of control and experimental mice. Spleen cells prepared from control mice exhibiting signs of severe GVHD were removed on day 5 to 6 and cultured overnight before using as responding cells in PLT assays. Similarly, spleen cells were prepared on days 7 to 8 and 25 to 26 from experimental animals showing signs indicative of long term survival. As presented in Table 38, cells from the GVH control mice responded against B6, BIO.AOR) and B10.A(5R) cells expressing the I-A derived gene product similar to the F^ hosts. In contrast, cells obtained from the long term surviving mice failed to respond to any of the stimulating cells tested. Similar results were found with cells removed from B6 host mice (Table 39). In this case, however all the cell populations were tested and found to be activated by the

PAGE 184

Table 38. The inability of lyitphocytes obtained from r^wborn CBA/J suppressed GVHD to respond in a PLT. 168 H2 Gene tics "^H-ldR Incorporation CPM+S) Control G7H CBA/J newborn Stimulator primed cells (a,b) suppr cells essed GVH Strain K A J E D (a,c) Day 7 Day25 none 829+ 115 3349+175 BIO. BR k k k k k 753+ 69 1206+ 36 2785+457 B6 b b b b b 24828+1195 3043+808 4239+ 85 BIO.MBR b k k k q 1111+ 171 1454+ 86 B10.A(5R) b b k k d 16920+ 267 5553+324 BIO.AOR) b b b k d 21088+1515 3373+378 BIO.CD d d b b b 5163+ 592 3295+500 B10.A(4R) k k b b b 1081+ 209 2979+ 92 B10.A(2R) k k k k b 981+ 9 2179+495 a) PLTs vere harvested at 48 hrs following a 12 hr pulse with 1 HCi of -^H-TdR. b) GVH reactivity was induced in sublethally irradiated (BIO x B10.BR)F, mice with BIO. BR splenocytes Oiich were cultured overnight prior to use. On day 5, the spleens were collected, made into a single cell suspension and cultured overnight prior to use. c) Suppressed GVH reactivity was induced in sublethally irradiated (BIO x B10.m)F, mice with BIO. BR splenocytes which were oocultured with CBA/J newborn splenocytes overnight prior to use. On day 5, the spleens were collected, mde into a single csll suspension and cultured overnight prior to use.

PAGE 185

169 S 05 di C i) (0 C -H Ci W 0) c .1 3 s •r^ ^ B CO -p U-l O 4J u ^^ i ?, 22 M O iu-i 0) C CO (1) tn u •H en C 0) in -4J •H o W CQ c o x: 4-) -^ o + o g 4J di^ I 4J 83 w c O £i 73 4J (C H -a H (C (N^vDinroroinr* •^ in vo on CN vo n iH rH +I+I+I+I+I+I+I+I VD iH -( (VJ 00 iH CO in •^Tl•-i o VD in r-\ as CO •-^ rH CM t-t ^ 0) e: ^ ^ in m ^ w r^
PAGE 186

170 mitogens Con A and LPS. By day 60, those mice which did not develop chronic GVHD responded to third party alloantigens as well as towards the donor cells, indicating that the donor cells had been rejected and that the host animal has recovered from the radiation. The cell populations isolated from the CBA/J newborn mediated GVHD suppressed B6 hosts engrafted with BIO. BR cells were also tested for their cytolytic activity in the CML assay (Table 40). Spleen cells from both the control and experimental hosts were placed in culture and stimulated with interleukin 2 for 7 days. Cells from the experimental mice showed large loss in numbers due to death. Furthermore, when examined in CML, these cells failed to kill either the B6 or BIO. BR target cells. In contrast, cells from the control animals undergoing severe GVHD not only expanded in the presence of the interleukin 2, but also proved capable of lysing B6 target cells, syngeneic with the host mice. Experimental animals which have survived considerably longer than the GVH control animals were tested for chimerism by determining the presence of H-2 antigens of the donor (H-2'^ or H-2^^) and/or the host (H-2^) The splenocytes from a sublethally irradiated B6 mouse were lysed only by the anti-H-2 alloantisera plus complement

PAGE 187

171 Table 40. The inability of cells obtained frcan a newborn CBA/J suppressed GVHD to respond in a CML. (a). % GVH Caitrol Specific Release CBA/J Newborn (b) suppressed Target cells 18:1 6:1 17:1 8:1 4:1 B6 Con A blasts BIO. BR Con A blasts 31 15 2 -12 -3 8 -4 1 -4 a) The splenocytes from (BIO x B10.BR)F^ mice reconstituted with BIO. BR splenocytes supplemented with CBA/J newborn splenocytes v>ere removed fran the animals on day 15 and were fed purified interleukin 2 every second day for 1 week. b) This test vs performed in a 6 hr assay. Tte effector to target ratios were: 17:1, 8:1 and 4:1.

PAGE 188

172 (96%). The cells from a B6 mouse reconstituted by the BIO. BR cells treated with the CBA/J newborn suppressor factor were killed by both the anti-H-2 antisera as well as the anti-H-2 antibody, 66% and 33% respectively. This evidence would suggest that the reconstituted animal was a chimera. Chimerism was also demonstrable in a B6 mouse reconstituted with B10.A(3R) splenocytes treated with cells from CBA/J newborn splenocytes. Here 73% of the lysable cells were killed by the anti-H-2'^ antisera plus complement, while 24% of the cells were killed by the anti-H-2 antibody plus complement. Since this anti-H-2^ antibody was a monoclonal antibody with the K^ specificity, only CBA/J derived cells (H-2^) were lysed and the donor cells from B10.A(3R) which are K^ positive were not lysed. One cell which constantly appeared to expand in interleukin 2 maintained cultures was a basophil/granulocyte like cell (Figure 41). Interestingly, this cell appeared to be clustered with lymphoid cells. From these functional studies (Tables 36 and 37) it appears that these mice have no anti-self reactivity and that the suppression due to the CBA/J suppressor cells is complete as judged by these two functional tests used.

PAGE 189

Figure 41. A single cell suspension prepared from a (310 x B10.BR)F, spleen 25 days after reconstitution with BlO.ER cells ana i^wborn CBA/J splenocytes. Ttese cells vere cultured for 10 days in the presence of interleulcin 2. Nuiterous large basophilic cells containing granules are seen. Ttese cells were present v*ien the single cell suspension was initially prepared from ti^ GVH suppressed aniiral. These cells were taken from an aniiial undergoing chronic G7HD.

PAGE 190

174

PAGE 191

175 V. Genetic Restrictions in the ability of CBA/J newborn cells to suppress lethal GVHD In light of the apparent nonspecific suppression of adult immune reactivity in vitro by newborn spleen cells, it was surprising to find that not all newborn spleen cells were capable of suppressing GVHD in the donor/recipient combinations, e.g., the BIO. BR, SWR or the (BIO. BR x SWR)F, combination (Figure 35). Examination of this question in several genetic situations now suggests that there are genetic restrictions governing the ability of newborn spleen cells to elicit a suppression of subsequent GVHD. This point is clearly illustrated with a few representative experiments presented in Table 41. The first restriction that becomes apparent is the requirement for histocompatibility (most likely at the 17 Chromosome) between the newborn suppressor cell population and the donor cell population. Thus, whereas CBA/J newborn cells suppress If adult cells expressing the H-2 haplotype, they are incapable of inhibiting adult cells of unrelated adult cells of unrelated H-2 haplotypes e.g., BlO.WB (H-2^^) hosts engrafted with B6 (H-2 ) cells, as shown in Table 41 set 1.

PAGE 192

176 Table 41. The genetic restrictions in the suppression of lethal GVHD by newborn ^leen cells in host animals reconstituted with allogeneic adult spleen cells. Reconstituting cslls Hosts Day of IVkDrtality: (a) SET 1 ~ B6 BIO.WB 5, 6, 7 [6] B6 + CBA/J Ifewborn cells 6, 7, 8 [7] SET 2 B10.A(5R) B6 5, 12, 20 [12] B10.A(5R) + CBA/J Newborn cells 5, 14, 38+, 38+ [24+] B10.A(3R) B6 12, 12, 12, 14, 30. [16] B10.A(3R) + CBA/J Newborn cells 14, 19, 20, 50+, 50+, 50+ [34+] B10.A(4R) B6 7, 11, 12 [10] B10.A(4R) + CBA/J Newborn cells 11, 11, 11, 11 [11] B10.A(2R) B6 4, 5, 10 [6] B10.A(2R) + CBA/J Newborn cells 5, 5, 5 [5] BIO.MBR B6 5, 6, 7, 10, 10, 10 [8] BIO.MBR + CBA/J feivborn cells 4, 6, 10, 10, 10, 10, 11, 19 [10] BIO.HTT B6 7, 7, 8 [7] BIO.HTT + CBA/J newborn cells 7, 7, 8 [7] SET 3 B10.D2 B6 10, 12, 13, 16 [13] B10.D2 + CBA/J Newborn cells 8, 8, 8, 10 [9] B10.D2 + BALB/c Newborn cells 10, 10, 10, 10 [10] B10.D2 + SBC Newborn cells 7, 14 [10] B10.D2 + SEA Newborn cells 26, 45, 46 [39+] B10.D2 + DBA/2 Newborn cells 10, 12, 25, 60+, 60+, 60+ [38+] SET 4 SWR B6 8, 9, 10, 10, 10 [10] SWR + SWR Newborn cells 9, 10, 10, 11, 11 [10] a) Underlined dates represent mice that have lived considerably longer than control G7H animals. The number within [ ] indicates the mean survival time of a given series of animals. Ths symbol + indicates that those animals vere still living after that day and did not appear to suffer from G7HD.

PAGE 193

177 The apparent H-2 restriction observed with CBA/J newborn cells using various strains carrying recombinant H-2 haplotypes derived from the a and b parental haplotypes were also investigated. As presented in Table 41 set 2, CBA/J newborn cells suppressed the GVH reactivity of BIO.AOR) and B10.A(5R) but not of B10.A(4R), B10.A(2R), BIO.MBR, B10.D2 and BIO.HTT. A second genetic factor in the suppression of adult GVH reactivity by newborn spleen cells apparently involves a non-H-2 genetic system, which at this time correlates with the expression of the Mis locus. In the examples thus far presented, CBA/J and AKR but not BIO. BR, SWR, B6 or (BIO. BR X SWR)Fj^ newborn spleen cells elicited suppression. Two further examples are presented in Table 41 sets 3 and 4. In the H-2 haplotype system, DBA/2 and SEA but not SEC or BALB/c newborn spleen cells were able to elicit suppression of adult B10.D2 reactivity in B6 hosts. In the H-2^ haplotype system, DBA/1 and BUB but not SWR newborn spleen cells suppressed GVH activity of adult BIO.Q or SWR cells in B6 host mice. Each of the strains which proved capable of eliciting a suppression expresses the strong Mis locus a/d phenotype, (Mis ), while all the strains which proved incapable of suppressing the GVH reaction express the null allele of Mis, Mls^.

PAGE 194

178 vi. The presence of newborn spleen cells incapable of suppressing the GVH reactivity of adult cells fails to modify the response of the sensitized donor cells Cells recovered from mice which have been engrafted with adult plus newborn cells but which still undergo severe GVH disease exhibit the same in vitro response pattern as cells recovered from the spleens of hosts dying from GVH induced by adult allogeneic cells alone. This is shown in Table 42 in which B6 hosts were engrafted with adult SWR cells or adult plus newborn SWR cells. The SWR cells proved incapable of eliciting suppression as they bear the null allele of Mis. The cells from each set of hosts when stimulated in vitro with a panel of secondary MLR stimulating cells exhibited identical response patterns. C. Attempts to prevent GVHD using newborn spleen supernates i. Characterization of the newborn supernate The newborn spleen associated suppressor cell population has been shown by Peeler et al. (106) to elicit its suppressor activity in part through secretion of soluble, culture stable materials, which in turn initiates activation of the suppressor limb of the immune response.

PAGE 195

179 Table 42. The Ability of lyirphocytes obtained fron newborn SWR splenocytes suppleniented GVHD to respond in a PLT. i H-TdR incorporation CPiVH-SD (a) Stimulator Cells SWR anti B6 SWR + SWR NB anti B6 (b) (c) 419+ 259 237+ 328 2718+ 210 2136+ 45 50668+9675 62687+12826 85819+ 81 84995+ 1704 57390+1446 61700+ 2250+2678 1767+ 1398 14405+3122 10627+ 2610 3646+ 296 1846+ 17 7169+ 257 5928+ none SWR B6 B10.A(5R) B10.A(3R) B10.A(4R) BIO.GD BIO.MBR BIO. BR > a) PLTs v^re harvested at 48 hrs following a 12 hr pulse with 1 nCi of H-TdR. b) G7H reactivity vas induced in sublethally irradiated B6 mice with 3 splenocytes which were cultured overnight prior to use. On day 5, the spleens vere collected, made into a single cell suspension and cultured overnight prior to use in PLT. c) G7H reactivity was induced in sublethally irradiated B6 mice with 3flR splenocytes which vere oocultured with SWR newborn splenocytes overnight prior to use. On day 5, the spleens were collected, made into a single oell suspension and cultured overnight prior to use. -4

PAGE 196

180 Before carrying out in vivo studies with newborn supernate it was decided to further characterize the factor (s) involved. Newborn supernates from various mice were collected and pooled. When 200 mis of the supernate was collected it was concentrated using an Amicon PM 10 filter. This filter allows the compounds with a molecular weight less than 10,000 daltons to pass through, while retaining those substances with a higher molecular weight. When a final volume of 10 mis of the >10,000 M.W. material was obtained, a dose response curve of suppressor activity by this preparation was determined (Figure 42). The dose response curve of this supernate illustrated a bimodal shape with a suppression of 30 to 50% over a range of 1.2% to 10% v/v. No genetic restrictions were seen when these in vitro tests were performed. As a control for this activity, 10% normal mouse serum was used. This was based on the fact that the newborn supernate was generated in 0.5% NMS and these supernates were concentrated 20 times the serum equivalent. When CBA/J newborn supernates were generated in serum free media a similar dose response curve was obtained. ii. Size profile of the newborn supernate factors Concentrated newborn supernate was passed through a sephacryl 300 column. The material was collected, each fraction was concentrated 10 fold, filter sterilized, and

PAGE 197

CQ (^ in j^ 3 V88 •i-t CO 4J 0) a a; O (0 OT >< 0) Vj XI 4-> o o g a> (1) ."J 8'^^8'y

PAGE 198

182 CO m < u O UI LlJ 2 cr Si 2cn z 3 >Q. H < L. if en Q 00 o Q O Q UJ Q Q < o !5 Z bJ O O ro O
PAGE 199

183 tested for suppressor activity as shown in Figure 43. Three peaks of activity could be found: one low molecular weight substance which eluted off in the last fraction and corresponded to the low molecular weight substance which passed through the Ami con PM 10 filter, a second fraction with a molecular weight similar to bovine albumin, or 60,000 daltons, and a third substance with a molecular weight higher than albumin, but less than the molecular weight of IgG. These compounds were not alpha fetoprotein because by Ouchterlony analysis no line of precipitation was observed, whereas, a line of precipitation was seen when AFP was used (Figure 44). No attempt was made to determine whether this larger suppressor factor was a dimer of the 60,000 dalton protein, due to the scarcity of the protein. Analysis using a SDS-acrylamide gel has exhibited 2 bands in the range of 6000 to 9000 daltons which are present in cell supernates derived from C3H/HeJ, BALB/c or BIO.M newborn mice. Neither band was present when the cells from adult mice were cultured for an identical time (Figure 45). In addition, a higher molecular weight species was identified in the 60,000 dalton region of the gel which corresponded to the higher molecular species isolated earlier.

PAGE 200

n) I •rH as. o 'O :^ (U P3 w Cfl .. &> o Ij M (C Q) >-( & ?] ^ -. R C C O (U 5 a 0) (1) iw X O •H ^ > Q 4J 03 a C +J 4J .H 3 -73 CO m „ •H tjl 4-) -H JJ nH (0 5 Cn c >-( -H -H (d 3 4J to (C to (C c ^ (U -H Vl s >-i 4J c o o (U O O 4-1 -H fP (0 -H O _^ IM U M-l (d (0 4J J-l 4-1 ^ •—I ii 8 o 0) iH •H &< O (0 n r-t Eh ^ >H (D O^CN >H (0 Q W MH m H !H rH t3 O 5 rH •H Eh C r-t U ro (1) a • >i 3 G 'l^ --* ^ (0 (U rH rH >1 X: -H (C >4 (DC |ii M -H > ro a — • m a >io '^ e p 4-1 O O -H CO -rH O O (d > 'O 4J ^ d) (d ^ ^ to Jh W M e 4-1 S-H o 4-) Lf) O tt) to fa Jh § rH en to ax; (tj c C 3 to 4J V4 to V4 (d (d 5 0) u Oi (d -H -H C c i2 j3 to -H

PAGE 201

185 o • > E (X UJ z g Io < p o c i Q. o a> ^ OZC/> O" > o o o o o o o o O o 0) 00 h in ^r ro e\j N0ISS3dddns lN30d3d

PAGE 202

r-1 8 5 4J I --" _^ o fl> 'd -H x: at vj n-i W (rt lu iH (D c 'OS H m (u o +J jC -H V4 (C H U-l 4-1 4J -H _^ QiCU 3 (0 w c JS Xi u o y -Q ,K (^ 9J S ^ O t3 >i C 0)
PAGE 203

187

PAGE 204

Figure 45. a)S-acrylamide analysis of the newborn factors. Splenocytes from either newborn C3H/He, BALB/c and BlO.M mice of adult C3H/He and BALB/c were incubated in PBS for three days. Afterwards the supernates were electrophoresed on a 15 to 30% continous SDS-acrylamide gel. The order of supernates tested were adult C3H/He (lane 1), newborn C3H/He (lane 2), newborn BlO.M (lane 3), newborn BALB/c (lane 4) and adult BALB/c ( lane 5 ) Fetal calf serum was a control in lane 6, v*iile different adenovirus proteins served as a control for the molecular v^eight markers (lane 7).

PAGE 205

189 J--28K 24 K -I2K i|-8.2K I I' I

PAGE 206

190 iii. Time of addition studies of the newborn factors Testing the fractions with the molecular weight of 60,000 daltons and less than 10,000 daltons, it was possible to examine the effect of each in the MLR with respect to time. The results of such a study are shown in Table 43. The higher molecular weight substance is capable of suppressing the reaction up to 48 hrs. In contrast, the low molecular weight substance does not exhibit any time dependency. The time of pulsing the primary reaction does not alter the results seen in Table 43. The results are identical when the culture is pulsed with ^H-TdR at 96 or 120 hrs. This illustrates that the newborn factors simply do not delay the primary response but do indeed suppress the reaction. Both of the newborn derived suppressor factors were tested for activity following a) heat treatment at 56 C for 90 min, b) UV light treatment, or c) incubation with 2.5 ng of trypsin for 1 hr at 37 C. The first two treatments did not affect the biological activity of either factor (data not shown); however, treatment with the trypsin altered the activities of both factors (Table 44). Another aspect of this experiment was an attempt to tentatively identify the nature of these suppressor factors. The MLR was subjected to treatment with either histamine or prostaglandin E2 to determine what, if any, effect these naturally occurring agents have on this reaction. Histamine

PAGE 207

Table 43. The suppression of primary MLR by adding rewborn supernate factors at different times. 191 H-TdR Incorporation CPM+SD (a) Time of addition Control MLR (b) 18 hours 24 hours 48 hours 72 hours 25078+1011 High MV added Low IVW added (c) (d) 14168+1616 12841+ 394 14225+ 591 17055+1616 23413+3383 15558+2553 17097+ 999 15530+ 41 18841+1498 14491+ 98 a) Primary MLRs v^re harvested at 96 hrs of the reaction following a 12 hr pulse with 1 nCi H-TdR. b) The primary MLR was BIO. BR splenocytes reacting against irradiated BIO ^lenocytes. c) The high molecular veight substance was added to the primary MLR culture at various time points. The final voluite of the factor was 10%. d) The low molecular veight substance was added to the primary MLR culture at various tine points. Tte final volutte of the factor was 25%.

PAGE 208

192 Table 44. The effect of various substances on a primary mixed lynphocyte reaction. addition none — control (a) Histamine 5.4 x 10~^M 5.4 X 10~3 5.4 X 10_^M 5.4 X 10 rM 5.4 X 10~'m Prostaglandin EL 10_!:m 10_iM 10_M 10 -^"m H-TdR Incorporation % of Control Response 22935+3502 20855+4078 35186+5284 38466+4183 27040+1569 23420+1246 9352+ 660 11457+ 853 18667+3665 20148+1128 25070+2979 CBA/J Newborn High ^W Factor 10189+ 796 CBA/J Newborn High MW Factor Treated with Trypsin (b) 32697+8435 CBA/J Newborn Low NW Factor 10743+3003 CBA/J Newborn Low m Factor Treated with Trypsin (c) 18275+3822 Trypsin (d) 22717+6235 100 91 153 168 118 102 41 50 81 88 109 44 143 47 80 99 a) A primary MLR vas established by BIO. BR splenocytes reacting ^gainst B6 splenocytes. The cultures were pulsed with 1 ^Ci of H-TdR at 96 hrs for 12 hrs. The various substances were added to the separate veils at the start of the reaction. b) A sanple of the high MW suppressor factor dsrived fran CBA/J newborn splenocytes was incubated with 2.5 ng of trypsin for 2 hrs at 37 C prior to use in this culture. c) A sanple of the low W suppressor factor derived from CBA/J newborn splenocytes was incubated with 2.5 ng of trypsin for 2 hrs at 37 C prior to use in this culture. d) An aliqout of the 2.5 ng of trypsin was incubated by itself in a volume eqiaal to that containing the newborn suppressor factors for 2 hrs at 37 C prior to use in this culture.

PAGE 209

193 m a concentration over a range of 5.4 x 10 M to 5.4 x -7 10 M demonstrated little if any inhibition of proliferation, but actually enhanced the proliferation at concentrations between 5.4 x 10~ and 5.4 x 10~^. Prostaglandin E however, showed a different dose response phenomenon. High doses of this substance (10~^ M -7 and 10 M) suppressed the reaction. It is unlikely that the newborn material is Prostraglandin E^ because these doses of prostaglandin suppressed iVlLR reactivity only if added during the first 48 hrs. Additionally, the molecular weight of the proteins seen in the SDS-acrylamide gel was much higher than the molecular weight of prostaglandin E iv. Effects of the supernate in GVH To determine if supernate material from newborn spleen cells can also suppress lethal GVHD (B6 x BALB/c)Fmice were sublethally irradiated and reconstituted with 40 x 10 BALB/c cells with or without newborn supernate. One half milliliter of the newborn supernate was injected i.v. on days 1,3 and 5. The results of one experiment are shown in Figure 46. Unfortunately, 9 of the 10 experimental hosts died. However, the fact that 1 experimental mouse survived long term may indicate the potential of newborn suppressor factor in preventing acute lethal GVHD.

PAGE 210

so '^ m ^ >i 0) -3 JO to iw m c o 3 r-H 4.) o a c 4J \ 0) c 09 jq CO B to tj Q C in T! a 3 o > so ^o3 S w fo

PAGE 211

195 o IT) o o LlI o to O CVJ < q: UJ cc UJ hu. < CO >< 4-T"' ^ X > O hT O o o in in nVAIAdnS lN3Dd3d

PAGE 212

196 In addition to using whole newborn supernate, the two isolated fractions were also tested for suppressor activity. Only the low molecular weight factor prolonged survival in the host animals where 1 of 6 experimental animals survived longer than 5 weeks. The high molecular weight material failed to prevent GVHD. In addition, treatment of the donor BIO. BR splenocytes with the suppressor factors plus adult CBA/J splenocytes failed to prevent GVHD in the experimental B6 mice. These experiments would suggest that the newborn suppressor cells may be more efficient than the secreted suppressor factors alone in preventing GVHD. D. Attempts to prevent lethal GVHD using AFP AFP has been shown to inhibit quite effectively in vitro cell mediated reactions (81,82). Since AFP is present in the sera of newborn mice, the question can be raised whether AFP will suppress GVHD. Pregnant mice were sacrificed between days 10 and 13 of fetal gestation. The amniotic fluid was aspirated into a collection flask. After 100 mis of the amniotic fluid was collected it was passed over an affinity chromatography column using rabbit antimouse AFP antibody. The antibody and affinity column were prepared by Dr. A. Kimura. The AFP which was eluted off the column was dialyzed and concentrated. This AFP was shown to

PAGE 213

197 bind to the rabbit anti-mouse AFP antibody (Figure 44) and demonstrated a line of identity with an AFP standard obtained from Dr. R. Murgita. This AFP was shown to suppress a primary MLR (Figure 42). GVHD suppression was attempted in 3 ways. The first method involved incubating virgin BIO. BR splenocytes overnight in the presence of a dose of AFP which suppresses the primary MLR (Figure 42). When these cells were injected into sublethally irradiated (BIO x B10.BR)F, mice, no GVHD suppression was observed (Table 45). The animals died on days 8,9 and 10, while the control GVHD mice died on days 7,9 and 11. The second approach was to generate primed cells from a MLR run in the presence of AFP. The concept here was to obtain specific suppressor cells generated by AFP which could subsequently inhibit GVHD. As in the previous approach, the treated mice failed to survive past day 11. Unfortunately, it was impossible to determine whether the mice died from GVHD induced by the donor allogeneic cells or the AFP primed cells which were injected. Thus, this approach was abandoned because of the technical problems. A third approach was attempted to determine whether systemic levels of AFP in pregnant females was sufficient to prevent acute lethal GVHD. In this protocol pregnant CBA/J mice were sublethally irradiated and reconstituted with 40 x

PAGE 214

198 Table 45. Lack of C57HD suppression in AFP treated (BIO x B10.BR)F, mice. Reaction: Day of Mcrtality: (a) Control Lethal GJBD (b) 7, 9, 11 [9] AFP treated mice (c) 8, 9, 10 [9] AFP treated BIO anti-BlO.BR prilled cells (d) 10, 10, 11 [10] a) Day in viiich the mouse died. b) BIO. BR splenocytes vepe cultured overnight in media. The following day 40 x 10 cells v;ere injected i.v. into sublethally irradiated (BIO x B10.BR)F, mice. Using the BIO splenocytes in place of BIO. BR cells generated an identical mortality rate. c) BIO. BR splenocytes were cultured overnight in AFP at a dose \n*iich vs shown to inhibit primary MLRs. The following day 40 X 10 cells vere injected i.v. into sublethally irradiated (BIO X B10.BR)F^ mice. d) BIO anti BIO. BR oells were generated in a primary I<1LR in the presence of AFP by Dr. A.B. Peck. Ttese prirted cells were incubated with virgin BIO splenocytes for 2 hrs before being injected into sublethally irradiated (BIO x B10.BR)F, mice.

PAGE 215

199 / 10 BIO splenocytes. As shown in Table 46, pregnant animals did not survive longer than non pregnant mice. Thus, this experiment suggests that AFP levels which exists in the third trimester of murine gestation were insufficient to prevent acute GVHD.

PAGE 216

Table 46. Inability of pregnant females to be suppressed fron acute GVHD. Condition: Day of Mortality: Non pregnant CBA/J (a) 15, 16 [16] Pregnant CBA/J (b) 15, 16, 21 [17] a) Two CBA/J fe^les were sublethally irradiated and reconstituted with 40 X 10 BIO ^lenocytes. Mortality was scored on tte day the iTDuse died. b) Three CBA/J females which were pregnant (days 14 to 15) were sublethally irradiated and reconstituted with 40 x 10 BIO splenocytes. Mortality was scored on the day the mouse died. 200

PAGE 217

DISCUSSION 4.1 Need For Immunosuppression For the Developm ent of GVHD The first set of experiments (Tables 2 and 3) revealed graft versus host reactions were only observed in immunocompromised individuals. Nonirradiated F^ mice, which might be expected to be susceptible to parental cell attack, did not develop GVHD, contrary to the classical laws of transplantation (see section 1.1). On the other hand, those animals which were sublethally irradiated and reconstituted with allogeneic cells developed classical GVH conditions such as wasting, diarrhea, lethargy and hypothermia. The dose of irradiation used to suppress the host and make GVHD possible was not sufficient to kill the host via total immunosuppression due to the associated leukocytopenia (Figure 4). The dose used in these studies, then, allows the host's immune system to recover to a degree which protects the host from its environment, yet weakens the host enough so that engraftment of allogenic cells are i':^

PAGE 218

202 generally lethal. This may approximate the situation in bone marrow transplants when bone marrow engraftment does occur. 4.2 The Kinetics of GVHR In general, GVH reactions followed distinct kinetics (Figure 5). Combinations which differed at the entire H-2 locus produced the mortality within the first 10 days. In the allogeneic reactions, death occured between days 5 to 10 while in semi-allogeneic reactions mortality was observed at days 8 to 14. In class I disparate combinations using K/D or D region differences, mortality occurred between days 10 to 25; however about 25% of the recipients survived longer than 5 months. In I region mismatches, mortality was dependent on the genetic combination. Complete mortality was observed by day 18 in BIO.AQR anti-(B10 .T( 6R) x B10.AQR)F, (Figure 5) and (BIO x B10.Q)F anti-BlO.MBR (Figure 23). In (BIO.MBR X B10.GD)F^ anti-BlO there was a 30% mortality (Figure 24), while in BlO.SOR) anti-(A.TL x B10.HTT)F no mortality was observed even after 2 months after cell engraftment (Figure 25). Minor histocompatibility mismatches also produced different survival patterns: BALB/c anti-DBA/2 exhibited little if any mortality, B10.D2 anti-DBA/2 demonstrated a more severe reaction with 40% mortality

PAGE 219

203 (Figure 26), CBA/Ca anti-AKR proved to be the most severe of the minor histoincompatible reactions with 80% mortality after 30 days (Figure 27). Thus a wide array of survival patterns were discovered and most interestingly, different functional activities were found in different reactions. 4.3 The Functional Activities of the T Lymphocytes Recovered from GVH Animals GVH reactions are thought to be initiated by T lymphocytes when these donor cells come into contact with the recipient's histocompatibility antigens. These T cells become activated against the foreign antigens and respond in a number of ways, i.e. by becoming cytotoxic T cells, helper T cells and/or suppressor T cells. I found that T cells generated in GVHR can be recovered with functional activities and these reactivities are highly specific. It has generally been assumed that the in vitro correlate of GVH is the mixed lymphocyte reaction (109). A great deal of information has been obtained using in vitro tests. Most of the reactions which occur in vitro do correlate with the in vivo work, therefore it can probably be inferred that the two situations do have similar factors responsible for reactivities which are observed. For example, it is known that cells which are primed to one antigen will only respond to that antigen in either primed

PAGE 220

lymphocyte tests or in cell mediated lympholysis reactions. Those antigens responsible for the initial stimulation of an immune response can then be readily identified. From the in vJ-tJ^o tests, it has been possible to test whether the reactions which occur in vitro will truly predict what its in vivo counterpart should do. The PLT assays have revealed that class II molecules are responsible for the vast majority of proliferation which is observed in vitro by the GVH primed cells. These observations also correlate well with previous studies by Wolters and Benner (110,111). In their work, GVH primed T cells from the spleen and lymph nodes, were capable of being adoptively transferred into normal virgin animals and then were capable of mounting delayed type hypersensitivity reactions against challenges of primary stimulating I region antigens, but not against K/D antigens. Several possibilities exist to explain this phenomenon. First, suppressor T cells may be present which inhibit K/D proliferation from occurring. This explanation can be eliminated because when GVH primed cells are cocultured with MLR primed cells, which are permissive for K/D proliferation, the resultant proliferation towards K/D antigens occurred (Table 7) illustrating that specific K/D suppressor cells are not present. Another possibility is that incubation of the GVH lymphocytes with contaminating

PAGE 221

205 PMNs may alter the responsiveness of the lymphoblasts To rule out this hypothesis, GVH primed lymphoblasts can be centrifuged with a higher density Ficoll-Hypaque (d=1.101) which will eliminate all contaminating PMNs and dying cells. When these cells are tested (Table 7) the same proliferative responses are again seen. Thus, it appears that this overall phenomenon is not artif actual, but is indeed real. It would argue that some unknown in vivo environmental influences are operating at this level. The filtration of K/D primed cells in the periphery before these cells arrive in the spleen or lymph nodes may be a possibility, but an unlikely one. The path of injected cells in the host is: blood, lung, lymph node and liver. Early studies by Ford and Gowans (112) have shown that by 12 hrs the majority of donor cells have arrived within the spleen and lymph nodes in the host. The absorption studies performed here indicate that in vitro primed proliferative cells do not attach to mouse fibroblast monolayers possessing the appropriate K/D antigens. This is in contrast to studies by Brondz et al (113) who showed that CTLs are capable of being absorbed onto fibroblast monolayers, possessing the appropriate K/D antigens. In addition, cytotoxic T cells directed at the K/D antigens are found in the spleen (Tables 20 and 21) and this would indicate these cells are already present within the spleen and have not

PAGE 222

been filtered out in the periphery. Thus, the lack of proliferative reactivity against the class I molecules must have a different basis. Another in vivo environmental effect is evident when a comparison between H-2 and I region GVH reactions are compared. Entire H-2 mismatched reactions produce many 6 fi primed cells, 2 x 10 to 5 x 10 cells/spleen, while I region or K/D mismatched reactions only yield about one sixth of the number of primed cells (Table 4). In addition, the quality of the primed cell reactivity in I region GVH is substantially weaker than the H-2 primed cells, despite the fact that the same number of cells were used to initiate the GVHR as well as used in the PLT assays. In contrast, the MLR system predicts both reactions will yield large numbers of primed cells which will respond very well in PLT assays. This anamolous finding indicates some in vivo synergistic effect is occurring in the entire H-2 GVHR, while in the I region GVHR an essential component is either diminished or is missing. The same type of finding is also observed in the Mis minor histocompatibility systems. In vitro reactions against the Mis determinants are very strong and demonstrate excellant secondary responsiveness, similar to I region mismatches. Primed cells generated in minor histocompatibility GVH reactions fail to show strong responsiveness (Tables 16 and 17).

PAGE 223

207 In K/D disparate reactions, no proliferative responses are demonstrated (Table 15), as would be predicted from the data observed in the entire H-2 mismatched reactions (Tables 6,7,10,11,12,13,14). For some unknown reason these GVH primed cells do not respond in the same way as do the in vitro primed cells. In primary MLRs K/D disparate combinations are very strong. It would appear that either the in vivo environment does not permit proliferation towards K/D antigens or that in vivo restraints placed upon these reactions are not present in the in vitro conditions. The former possibility is probably correct because if the GVH primed cells derived from an entire H-2 mismatch are restimulated twice in vitro the pattern of restimulation that is observed is identical to that pattern seen when the GVH primed cells are first tested. No additional reactivity is generated against the class I molecules, indicating no new antigens are recognized de novo What the mechanism of restraint is, remains only speculative. Perhaps, the crowded environment of the spleen, lymph node or liver does not allow the lymphoblasts to expand to the same extent as they do in the free environment of a tissue culture flask. In the flask conditions, the primed lymphocytes probably keep expanding until optimal or maximum cell concentrations are reached. It should be noted

PAGE 224

208 that the primed cells obtained from GVH animals on days 5 to 14 of the reaction appear to be in a primed state and do not show any residual mitotic activity (as determined by H-TdR incorporation by primed cells alone). When MLR reactive cells are tested too early without allowing sufficient time for the primed cells to revert back into small lymphocytes, background H-TdR incorporation is very high and frequently masks any secondary responses which may be present. From this observation it is postulated that the GVH primed cells have matured quicker and have reached an equilibrium. These endstage cells do not need to replicate anymore and now these cells only mediate their effects without further need for cell division. This explanation could explain why proliferative responses to K/D antigens are not seen. In vivo the CTL precursors proliferate quickly in response to K/D antigens and this intermediate step is greatly shortened or lost, while in MLR conditions this intermediary stage has been prolonged indefinitely due to the lack of the in vivo restraints. Cytotoxic T cells are also readily apparent in entire H-2 disparate reactions (Tables 20, 21), supporting previously published data (114). The presence of CTLs in tissue other than lymphoid e.g. liver (Table 21), is a new finding and does fit with the current dogma concerning GVHD

PAGE 225

;. 209 pathogenesis. Although CTLs are seen transiently in K/D disparate reactions, present in the spleens on day 7 of the GVHR but not by day 22, no CTL activity is recoverable when (Table 22) these animals are dying with GVHD (Figure 5). The presence of transient CTL activity does agree with a previous study by Hamilton and coworkers (115,116), who have suggested that CTL precursors do emigrate from the lymphoid tissue when they become active CTLs. Their studies showed that CTL precursors are present in the spleens of animals who are not displaying clinical signs of GVHD. However, when GVHD does start presenting itself clinically, CTL activity is readily apparent. Little if any CTL activity is detectable in I region GVHs (Table 25), but was demonstrated in the minor histocompatibility reaction of CBA/Ca anti-AKR (Table 32). When purified exogenous interleukin 2 was added in vivo to a K/D disparate combination, the mortality rate sharply increased and this rate approached that of an entire H-2 mismatch GVH (Figures 5 and 20). Interestingly, CTL activity in the spleens of these animals did not increase to the extent as would be predicted by CTL activity found in the entire H-2 disparate GVH. Instead, CTL activity that is observed is identical to the amount of CTL activity found in the normal K/D disparate GVHR (Tables 23 and 24). These data

PAGE 226

210 support Hamiliton and Parkman's premise that mature CTLs could have migrated out of the spleen. These IL 2 studies do not prove that CTL activity is responsible for the mortality that is observed. Other possible mechanisms of IL 2 activity exist such as: increased natural killer cell activity, increased helper T cell activity or activation of other leukocytes which trigger an inflammatory response. IL 2 has been reported to enhance NK activity in vitro (117) as well as help maintain helper T cells in vitro both in human and murine systems (118,119). Soluble T cell derived factors, possibly IL 2 are known to cause mast cells to degranulate (120), leading to a histamine release which is an early part of the hypersenitivity response. The presence of undetectable CTLs or CTL precursors which are expanded in vitro by IL 2 does seem to correlate with mortality fairly well. In an H-2 mismatch, the presence of large numbers of CTLs in the spleen and liver correlates well with the rapid demise of the hosts by days 5 to 14. The large number of CTLs no doubt plays a role in the host's death. In the BlO.AQR anti-(B10 .T( 6R) x B10.AQR)F, combination these expanded lines lyse the host's I-a'^ positive cells, but the vast majority of the cytotoxic T cells appear to recognize the donor cells as being foreign (Tables 26 and 28). Thus, a host versus graft reaction also

PAGE 227

211 appears to be a possible mechanism of the GVH process. Even though these hosts are nearly lethally irradiated {650r) the host is still able to mount some type of counter reaction. It should be noted that studies performed by Dr. A. Kimura (personal communication) have shown that even in 550r sublethally irradiated mice some form of cytotoxic cells can be generated. The histopathological lesions of GVH are identical to HVG (104), so the pathology does not reveal any clear distinctions. By the very nature of this reaction it is impossible to distinguish whether the donor or host cells are becoming the effector cells. Since both strains of mice possess the identical K^ and D molecules it would be impossible to distinguish donor cell from host cell by normal serological tests. In addition, the activated cells are T cells and they would not be expected to bear la markers in sufficient quantity to distinguish it from the other. Even if la markers were detected by very sensitive methods such as flow cytometry, the intrepretation would be questionable because Delovitch et_al. (121) have reported that donor T cells are capable of absorbing the host's la molecules during GVH reactions. One interesting finding of the previous combination is that the reaction of BlO.AQR anti-BlO .T( 6R) is very weak in its ability to generate CTLs or CTLp, while the reverse reaction, B10.T(6R) anti-BlO.AQR is a very strong reaction.

PAGE 228

,..212 In fact, this reaction is even able to overcome a sublethal dose of radiation to develop CTLs. This observation sheds light on a previously reported anomaly first described by Elkins (122), which was not readily explainable at the time its publication. In this system, Elkins demonstrated that when newborn BIO.AQR mice were injected with B10.T(6R) cells the majority of these mice died, while in the reverse direction, B10.T{6R) mice completely survived an injection of BIO.AQR cells. Since the newborn mice are totally immunoincompetent at birth and are unable to mount any kind of reaction, only GVH reactions occur. Thus, only those reactions in the hosts which develop sufficient numbers of CTLs or CTLp will develop mortality. From the polarity of the reaction in developing CTL activity in vitro with IL 2 it is apparent that B10.T(6R) cells should readily develop cytotoxic activity against BIO.AQR, but the reverse reaction is significantly weaker. Thus, Elkin's data can now be explained in terms of CTL activity in vivo In another I region mismatch, (BIO x B10.Q)F, antiBIO.MBR rapid mortality was observed (Figure 23). The time in which mortality was observed was identical to that corresponding to a whole H-2 mismatch (BIO x B10.Q)F anti-BlO.BR. When these GVH splenic cells are expanded with IL 2 in vitro for 10 days, both GVH and HVG reactions are present with about equal cytotoxic strengths (Table 29).

PAGE 229

.-213 This case is slightly different from the previous situation, where HVG was considerably stronger than GVH. In contrast to the two previous I region combinations, the (BIO.MBR x B10.GD)F anti-BlO combination produces a survival curve with 30% mortality (Figure 24), which indicates a weak GVHR. When these splenocytes from day 5 of the GVHR are expanded in vitro with IL 2, only a weak cytolytic response can be generated against the host (Table 30). Such a weak cytotoxic response could conceivably give rise to a weak GVHR. From the strength of this in vitro response it would appear that proliferative types of cells outnumber the CTL precursors and thus limit the amount of IL 2 that the CTLp could absorb. It would therefore be of importance to clone these cells so that a homogenous cell line could be obtained to demonstrate strong cytolytic activity. In the BIO.SOR) anti-(A.TL x B10.HTT)F combination, no CTL or CTLp activity is demonstrable (Table 34). Furthermore, no mortality is detected even after 40 days after engraftment (Figure 25). This shows that the lack of CTL activity matches the lack of GVHD mortality. In the minor histocompatibility combinations examined, cytotoxic T cells are demonstrated in those combinations in which mortality is seen. In CBA/Ca anti-AKR, cytotoxic T cells can be found which lyse AKR cells strongly, and DBA/1

PAGE 230

2U and DBA/2 cells weakly (Table 32). The survival curves of such reactions reveal incomplete mortality (Figure 27). In another combination, BALB/c anti-DBA/2, no cytotoxic effectors can be found in the animals on the days they were tested (Tables 33 and 34). But if cells from a BALB/c mutant strain, BALB/c (a strain which lacks the L molecule), are injected into DBA/2 mice, a completely different survival profile is obtained (Figure 26). This curve is one in which death occurs by day 12. Cytotoxic cells directed at the L chain have been reported (123) previously and it appears that in this GVH reaction CTLs may again play a major role in GVHD pathogenesis. In summary, these series of experiments provide a correlation between GVH mortality and the presence of cytotoxic T cell activity. These data tend to favor previous theories about in vivo graft rejection, whether it be skin, kidney or graft versus host; indicating that cytotoxic T cells play a role in tissue destruction. Recently, these theories have come under attack by several groups including McKenzie et al (124,125) and Mason (126,127). Their major arguments have relied on the phenotype of the cells recovered from graft rejection sites along with the corresponding negative CML assays. Because they have eliminated Lyt 2+ cells from their initial donor populations and fail to find subsequently Lyt 2+ cells or cytotoxic

PAGE 231

215 activity, they believe that no CTLs are present in their systems. In countering such studies it must be pointed out that cytotoxic cells do not necessarily have to be Lyt 2+. Swain et al (47 ) have shown that killer cells towards I region antigens are Lyt 1+ 2cells. In the present study the BIO.AQR anti(BIO .T( 6R) x B10.AQR)F^ effector cells generated were largely Lyt 1+ (80%) while some cells are weakly Lyt 2+ (15 to 20%). Furthermore, McKenzie and coworkers did not expand their recovered cell populations in vitro with IL 2 as done here and firmly rule out cytotoxic T cell potential. The data presented here, then, would tend to agree with the findings that when GVH primed cells are tested in CML assays the same day as they are removed from the host, there does not appear to be any anti host CTL activity. But what if McKenzie and fellow workers had expanded their recovered cells with IL 2? These studies simply correlate GVHD mortality with the presence of CTL or CTLp. They do not show the mechanism through which they are capable of working. The CTL or CTLp may expand and attack the host's leukocytes or bone marrow cells leaving the host temporarily immunoincompetent and thereby susceptible to infections. In H-2 disparate GVH combinations the large number of CTLs probably kill the epithelial cells lining the gut and the hepatocytes which leads to quick mortality. However, in combinations with

PAGE 232

216 slower mortalities the CTLs probably do not attack the intestine and liver, but are more active against target cells with high amounts of K/D antigens, such as leukocytes. 4.4 The Nonspecific Factors Influencing Mortality in GVH Besides the specific functional immunological reactions which have been discussed, there also exists nonspecific factors which could play a significant role in the pathogenesis of GVHD. Inflammatory cells most noticeably the polymorphonuclear leukocytes were always obtained from effected GVH tissue, i.e. spleen, lymph nodes and liver. Fifty percent of the recovered leukocytes in the entire H-2 GVH organs were PMNs, while in K/D or I region GVHD, 90% of these populations were PMNs. The presence of these cells could be explained two ways. First, the natural occurrence of PMNs: is that hematopoeisis in the mouse can occur in the spleen while the liver can be a site of extramedullary hematopoeisis (128). Second, the activated lymphocytes could be secreting chemotactic factors or growth factors, such as interleukin 3 ( IL 3). These factors would attract or stimulate colonies of granulocytes to develop in situ IL 3 has been reported to be secreted by T cells/cytotoxic T cells (129). In addition, IL 3 has been reported to contain granulocyte colony stimulating factors (130) so this

PAGE 233

217 possibility of PMNs being attracted to the GVH effected organs is indeed feasible. Irradiated control mice do show the presence of granulocytes, but the quantities of these cells (0.05 X 10 cells/spleen) is never as great as those seen in the GVH reactions (5.0 x 10 cells/spleen). Also, the lymph nodes are not sites of hematopoeisis, yet PMNs are found here, thus, the second possibility could be responsible for this phenomenon in the GVH lymph nodes. What ever the reason PMNs are present, the question still remains: do the PMNs play any role in the associated tissue destruction? From the 24 and 48 hr CML assays, it would appear that PMNs are not directly lytic, despite the prolonged CML assays (Table 34). However, from the liver histology, it is noticeable that the PMNs frequently have invaded the parenchyma along with the lymphocytes. The PMNs could be attracted into such areas by the T cells. Because the PMNs have such a short lifespan, these cells may not exit from this sites, but die and release their lysosomal enzymes. These enzymes could then exacerbate and accelerate any tissue destruction that is actively occurring. This model would support McKenzie's idea of a delayed hypersensitivity response occurring in graft rejection sites (124,125). This mechanism also tends to be supported by the indirect evidence seen in the human clinical situation, that when GVHD occurs it is frequently alleviated with

PAGE 234

218 inununosuppressi ve drugs such as cortisone and steroids, which are also known to inhibit inflammatory reactions (131). Another apparently nonspecific mechanism important in GVHD pathogenesis is depicted when primed T lymphocytes induce mortality in sublethally irradiated syngeneic, allogeneic or third party hosts. The pathology of these lesions in syngeneic mice are markedly different from the typical GVH reaction. The spleens of these animals are often packed with lymphocytes (Figure 33), while the livers of such animals are often congested and edematous. Isolated patches of cellular infiltrates with associated mild necrosis are found. Although there is no perivascular cuffing of leukocytes, these infiltrates are markedly distinct (Figures 34 and 35) from those seen in a normal GVHR (Figure 10). In contrast, the lesions found in allogeneic animals reconstituted with primed lymphocytes are similar to classical GVHD (Figure 29). The spleens are necrotic, the livers have infiltrates in them with associated coagulative type necrosis. It has been reported that donor T cells circulate in the course of GVHD. Studies by Sprent and Miller (132,133,134,135) have shown that donor cells on days 1 and 2 are primarily found in the spleen and lymph nodes of the host and that by day 4 blast cells can be collected from the

PAGE 235

219 immunosuppressive drugs such as cortisone and steroids, which are also known to inhibit inflammatory reactions (131). Another apparently nonspecific mechanism important in GVHD pathogenesis is depicted when primed T lymphocytes induce mortality in sublethally irradiated syngeneic, allogeneic or third party hosts. The pathology of these lesions in syngeneic mice are markedly different from the typical GVH reaction. The spleens of these animals are often packed with lymphocytes (Figure 33), while the livers of such animals are often congested and edematous. Isolated patches of cellular infiltrates with associated mild necrosis are found. Although there is no perivascular cuffing of leukocytes, these infiltrates are markedly distinct (Figures 34 and 35) from those seen in a normal GVHR (Figure 10). In contrast, the lesions found in allogeneic animals reconstituted with primed lymphocytes are similar to classical GVHD (Figure 29). The spleens are necrotic, the livers have infiltrates in them with associated coagulative type necrosis. It has been reported that donor T cells circulate in the course of GVHD. Studies by Sprent and Miller (132,133,134,135) have shown that donor cells on days 1 and 2 are primarily found in the spleen and lymph nodes of the host and that by day 4 blast cells can be collected from the

PAGE 236

220 thoraic duct. This finding demonstrates that primed cells traffic through the host during the course of GVHD and are not merely sessile. The ability of primed cells to mediate GVHD has also been observed in this study (Figures 28 and 36). These cells were primed against the host's histocompatibility antigens, prior to engraftment, and as such, probably are responsible for the tissue destruction that is seen. However, this assumption may not be correct, since primed helper T cell clones and lines known not to possess any cytotoxic potential are also capable of precipitating GVHD (Figure 36). Again the lesions appear similar to what is seen in GVHD caused by injecting virgin allogeneic splenocytes. The helper clones and lines are directed at the I-A^ molecule and this reaction is analogous to the BIO.AQR anti(B10.T(6R) X B10.AQR)F^ reaction. The mortality induced by the clones may be due to the proliferation of clones and associated graft versus host reaction via a hypersensitivity reaction. Evidence favoring this explanation is that a heteroclitic clone (clone 4 RY1/Y2) is least effective in generating mortality. The low responsiveness of this clone in vitro towards the I-A^ molecule could conceivably delay development of GVHD. While those strongly in vitro proliferative clones would mediate quicker GVHD. The other possible explanation is that the host may be responding

PAGE 237

321 against the graft as was observed in the BIO.AQR anti(B10.T(6R) X B10.AQR)F^ reaction previously. Finally, a third possibility could be that the donor cells respond against the host and produce the necessary interleukins to maintain the host cells responding against the graft. Thus, it is still premature to conclude which reaction GVH or HVG is directly responsible for the mortality which is seen in this situation. 4.5 The Attempts to Prevent GVHD Using Anti-Host I-A Antibody From the PLT of GVH primed cells it has been demonstrated that the I-A molecule is responsible for the major source of donor cell proliferation which is observed in vitro Experiments using anti host I-A antisera to inhibit subsequent recognition of the host's I-A antigen, thereby preventing donor cell recognition proved unsuccessful in this model. Possible explanations for this failure include 1) not all the antigenic sites were covered, and 2) the antibody was internalized and removed from the cell surface allowing the donor lymphocytes to recognize the host's I-A molecules Obviously different protocols need to be studied to

PAGE 238

222 determine the efficacy of using anti-I-A antisera. This approach was not pursued. 4.6 The Attempts to Prevent GVHD using Newborn Suppressor Cells ~ ^ From the work of Peeler et al (106) a suppressor cell and its associated factor have been described which operates by inhibiting in vitro MLR and CML reactions. It was therefore of interest to determine whether these suppressor entities would function in vivo by preventing GVHD. CBA/J (H-2 ) newborn splenocytes are removed from one to four day old neonates, have the capacity to prevent GVHD by donor BIO. BR (H-2^) cells in (BIO x B10.BR)F mice (Tables 35 and 36). This was not so when adult CBA/J splenocytes are cultured with the BIO. BR cells, which resulted in no inhibition of GVHD. Thus, the ability of CBA/J newborn cells to exert some type of GVHD suppression eliminates the possibility that the BIO. BR cells had simply become primed towards CBA/J determinants and therefore lacked the ability to respond towards the H-2'^ antigens. Those animals which received the BIO. BR cells treated with or without the CBA/J adult cells developed the classical GVHD syndrome: diarrhea, weight loss, lethargy etc. within the normal time frame. Most of the animals which received the BIO. BR cells treated with the CBA/J newborn

PAGE 239

isaij-': 223 cells did not show signs of GVHD at the time when the control host animals were dying. When these treated mice were dissected at day 15, leukocytic infiltrates were common in the liver. Perivascular cuffing was also prominent. The parenchyma was however intact and presented no cellular destruction, totally unlike previous GVHR where leukocytic infiltrates proved destructive to the parenchyma. The intestines were also affected, but not to the extent that was observed in the control GVHR. About one fourth of the villi were dilated and devoid of leukocytes, the rest of the villi appeared normal. In control GVH animals, nearly 95% of the villi were effected. Thus, the experimental animals did not appear to be wasting by the associated malabsorption problems. The leukocytes recovered from the spleen proved to be equivalent with the number of leukocytes recovered from the GVHD suppressed mice demonstrated little or no functional activity as compared to the reactivities seen by normal GVH primed cells. As these acute GVHD suppressed reactions proceed, the animals begin to degenerate by day 25. The mice begin to develop skin lesions: loss of hair and redness of the skin being common. Mononuclear cell infiltrates are found in the dermis (Figure 38). The liver also appears to be effected. Histologically the liver is completely destroyed, the parenchyma is "punched out", only nuclear remnants are left

PAGE 240

224 (Figures 39 and 40). No cell boundaries are visibly detectable. However, the intestines appear normal. When the spleens are removed, the number of recovered leukocytes was the same as day matched irradiated control mice, and these cells demonstrated no proliferative activity towards the antigens syngeneic to the host (Table 38). Furthermore when these cells are expanded with IL 2 for one week in vitro these cells exhibited no CTL activity (Table 40). Instead, a granulocyte/basophil like cell appear in culture (Figure 41). The presence of these cells was never experienced before using the IL 2 in vitro expansion technique. The presence of these cells was also noted when the spleens were freshly prepared, so it does not appear such cells developed de novo from other cells. Whether these cells are responsible for the chronic GVHD seen is not apparent at this time. ^ To determine if the CBA/J newborn cells had incapitated the adult BIO. BR cells, the BIO. BR cells were tested for mitogenic responses using Con A and LPS (Table 37). The newborn cells themselves were unable to respond to Con A and LPS. The BIO. BR cells treated overnight with the newborn CBA/J cells were still quite capable of responding to the T and B cell mitogens. This experiment establishes that the BIO. BR cells treated with the CBA/J newborn cells have not been killed or have been functionally inactivated.

PAGE 241

225 There does appear to be genetic restrictions in what types of GVHD the CBA/J newborn cells are capable of suppressing. When B6 splenocytes were incubated with the newborn CBA/J cells, these treated cells were not capable of suppressing GVHD against BIO.WB, another entire H-2 mismatched reaction (Table 41). Neither of these two mouse strains possess any of the histocompatibility antigens derived from the H-2 haplotype. This might suggest that the donor cells must share I regions with the suppressing CBA/J for GVHD suppression to occur. Another experiment (Table 41) using CBA/J newborn cells demonstrated that B10.A(5R) and BlO.AOR) cells are also capable of being suppressed enough to prevent acute GVHD in sublethally irradiated BIO mice. This experiment therefore correlates the I-E molecule with the CBA/J mediated GVHD ]r suppression. Whereas, a shared I-A molecule with the CBA/J cells does not appear to be sufficient for this suppression to occur. Unfortunately, this hypothesis does not hold when other mouse strains are studied. For example, B10.A(2R), BIO.MBR, and BIO.HTT which possess the I-E^ were not capable of being suppressed by the CBA/J newborn cells. Therefore, this restriction is unable to be explained at the present time. Another study demonstrated that newborn splenocyte mediated GVHD suppression can occur in the H-2 system

PAGE 242

226 (Table 41). In this combination newborn mice from either DBA/2 or SEA were capable of suppressing B10.D2 splenocytes from causing acute GVHD in B6 mice. Quite unexpectedly newborn splenocytes from BALB/c and SEC mice failed to suppress GVHD from occurring. To determine whether any newborn spleen cell preparation would demonstrate GVHD suppression, newborn (BIO. BR X SWR)Fsplenocytes were used in place of the CBA/J newborn cells. As demonstrated in Table 35 no suppression of GVHD in (BIO x B10.BR)F mice was induced, despite the fact that the same protocols and techniques were employed. To eliminate the possibility that the H-2^ haplotype interferred with the priming reaction newborn (BIO. BR X SWR)F^ cells were incubated with (BIO. BR x SWR)Fj^ adult cells overnight and then injected into (BIO x B10.BR)F^ mice. Again the same negative results were observed: no GVHD suppression was observed. These previous studies have indicated that cellular interactions, most probably within the I region are needed before GVHD suppression can be effected. When newborn spleen cell supernates are used they do in fact inhibit MLR primary reactions irregardless of the genetic disparity. And these factors only marginally prevent GVHD from occuring in (BALB/c x B6)F^ mice reconstituted with BALB/c cells

PAGE 243

227 (Figure 46). This argues that some unknown cellular interactions are probably responsible for GVHD suppression. The newborn suppressor factors can be fractionated into 2 populations: one with a molecular weight less than 10,000 daltons and the other with a molecular weight greater than 10,000 daltons. The mode of action of these substances appears to be different. The time of addition study revealed that addition of the low molecular weight substance could inhibit a primary MLR even up to 72 hrs, right before the 3 H-TdR was added to the culture. This low molecular weight substance probably contains either: histamine, thymidine or prostaglandins which simply reduce "^H-TdR incorporation nonspecif ically. However, histamine's activities could be eliminated because it was unable to inhibit thymidine incorporation in mixed lymphocyte reactions. Prostaglandin E^ did suppress the mixed lymphocyte reactions, but this substance was only effective if it was added within the first 48 hours of the reaction, so this factor is not likely to be the low molecular weight factor. In contrast, the higher molecular weight substance only suppresses the primary MLR within the first 48 hours of the reaction. The size profile of the suppressor factor reveals that the higher molecular weight substance probably contains 2 entities (Figure 42); one with a higher molecular weight

PAGE 244

228 a molecular weight roughly the size of albumin. This substance is not alpha fetoprotein, because when anti-alpha fetoprotein antibody is tested in a gel diffusion test, no lines of identity form with purified alpha fetoprotein (Figure 44 ) Previous works (88,89,13 6) have used newborn suppressor cells to inhibit local GVH assays in adult F^ animals as well as temporary suppression of systemic GVHD in 5 to 7 day old mice. Interestingly, the newborn splenocytes which were successfully used in such assays have come from CBA, AKR and A/J mice. These reports have shown that neonatal spleen cells or newborn splenocytes from 1 to 3 day old mice have the ability to suppress H-2 mismatched GVH reactions as measured by footpad swelling or splenomegaly methods. However these suppressor cells are short lived, since mice older than 5 days do not possess suppressor activity. Other studies have demonstrated that newborn splenocytes secrete suppressor factors which inhibit MLRs, CTL development and T cell mediated antibody synthesis (137,138,139,140). These workers also claim that suppressor factor synthesis declines in the newborn mice around days 4 to 7 like the GVH suppression observed by Ptak and Skowron-Cendrzak (88,89). Research by Arygris (141) has claimed that the newborn suppressor factors can be distinguished into 2 fractions.

PAGE 245

229 St.. one with a molecular weight less than 10,000 daltons and the other factor with a molecular weight greater than 10,000 daltons. However, the higher molecular weight factor was thought to be simply dimers or multiples of the lower weight fragments. These factors were also thought to be secreted by newborn T cells. In contrast, previous work by Peck and fellow workers (106,139) have associated the source of these factors to monocytes and mast cells. Whereas, this work does not address this particular controversy, the findings reported here confirm that newborn splenocytes secrete soluble factors which are capable of suppressing primary MLRs (142,143). Additionally, this work also demonstrates that newborn suppressor cells are capable of suppressing acute lethal GVHD in adult mice. One point worth noting has been the correlation of suppression with Mis positive mice. The exact relationship and role of the Mis positive cells remains only speculative. First, the Mis locus or a gene linked to the Mis may control the type or the amount of suppressor f actor (s) that are secreted. Newborn cells constantly secrete factors and this continous supply of factor provides the stimulus for GVHD suppression. Second, recognition of the Mis antigen does not lead to any harmful effect, but activates the suppressor limb of the immune system. Previous

PAGE 246

230 work by Festenstein (100) has shown that the Mis antigens do not behave the same way as the H-2I antigens do. For example Table 1 illustrates that I region antigens induce good splenomegaly and produce good stimulation indices, nevertheless, Festenstein has shown that Mis disparate combinations do not induce splenomegaly but instead induce good in vitro reactivity. Cytotoxic T cells and specific antisera have never been produced against the Mis antigen, despite numerous unreported attempts. Follow up studies have shown that in vivo CTL generating reactions involving H-2 and Mis disparate reactions produce fewer CTLs than do the comparable H-2 disparate reactions (144). Likewise, Arygris (145) has found that when DBA/2 cells or P-815 cells (a mastocytoma derived from DBA/2) are injected into B6 mice, suppressor cells and suppressor factors are induced both in vitro and in vivo These suppressor entities are able to inhibit MLRs and CTL development. Similar unreported findings by Peck testify to the uniqueness of this Mis disparate reaction. Finally, in a different experimental setting. Click has reported that Mis disparate marrow grafts survive much better even in H-2 mismatches (146).

PAGE 247

231 Whatever the mechanism of suppression is, one thing that remains is that adult mice reconstituted with allogeneic cells treated with CBA/J and SEA newborn splenocytes do develop chronic GVHD. With the exception of Rappaport et al 's combination: B10.D2 anti-( DBA/2 x B10.D2)F, (96), no reported study examined to date has described chronic GVHD in the mouse system. Sprent and Korngold have recently speculated that both chronic and acute GVHD are caused by T cells (147,148,149). The evidence supporting the cause of acute GVHD is fairly solid, in that treatment of lymphocytes with anti-Thy antisera plus complement eliminates all types of (3VH which would have appeared, both local and systemic. The evidence linking chronic GVHD is not so well delineated. This research has showed that T cell reactivity may be completely abolished (Tables 38 and 40) or working in such low basal levels that the normal acute GVHD effector mechanisms may not be working properly and that another route of GVHD effectors may be working in its place (Figure 47). These other effectors do not have to be lymphocytes, perhaps the inflammatory cells such as PMNs and monocytes may be the cells mediating this form of the disease. If this other effector limb is indeed working then the lesions which are observed do not have to be identical to those produced by the T cell mediated effector mechanism(s) Thus, this disease process which is

PAGE 248

.^. 232 observed here could then be more comparable to the situation in the human bone marrow transplantation setting where chronic GVHD occurs. Thus, this proposed model would also explain why a form of GVHD occurs in the humans undergoing bone marrow transplantation, even after the donor T cells have been eliminated from the reconstituting cell populations.

PAGE 249

Figure 45. Proposed model for GVH morbidity.

PAGE 250

234 ALTERNATIVE PATHWAY OTHER CELLS (^NK J) (^MfT) — (m^ (^^UNKNOWN EFFECTOR MECHANISMS ON OTHER TISSUES ANTIGEN SPECIFIC MAIN PATHWAY IFNMIF MAF^ DIRECT ATTACK ON HOST TISSUE i.e. LEUKOCYTES BONE MARROW OTHER KEY TISSUES CHRONIC GVHD ACUTE GVHD \ / DEPRESSED IMMUNITY ; INFECTIONS i DEATH

PAGE 251

REFERENCES 1. Elkins, W.L. Progress in Allergy. 15:78. 1971. 2. Grebe, S.C. and Streilein, J.W. Adv. Immuno l. 22; 118 1976. 3. Klein, J. Biology of the Mouse Histocompatibilty Complex. Springer-Verlag. New York. 1975. 4. Hood, L.E., Weissman,I.L. and Wood, W.B. Immunology. Benjamin/Cummings. Reading, Mass. 1978. 5. Hellstrom, I., Hellstrom, K. and Allison, A. Nature. 230:49. 1971. 6. Cohen I. and Wekerle H. in Autoimmunity, editor: N. Talal Academic Press. New York. 1977. 7. Gershon, R., Cohen, P. and Liebikaber, S. J. Immunol. 105:586. 1971. 8. Mills, V. Orbach-Arbouys, S. and Gershon, R. J Immunol. 114:200. 1975. 9. Cerrotini, J., Nordin, A. and Brunner, K.T. J Exp. Med. 134:553. 1971. 10. Goldstein, P., Wigzell, H., Blomgren, H. and Snedmyr, E. J. Exp. Med. 135:890. 1972. 11. Emerson, E.E. and Thursh, D. J. Exp. Med. 137:1293. 1973. 12. Atkins, R.C. and Ford, W.L. J. Exp. Med. 141:664. 1975. 13. Ford, W.L., Simmonds, S. and Atkins, R.C. J. Exp. Med. 141:681. 1975. 14. Rolstad, B. Transplantation. 21:117. 1976. 235

PAGE 252

236 15. Lankford, J., Blackstock, R., Skyatalowicz V. and Hyde, R. Cell Immunol. 14:163. 1974. 16. Katz, D. Lymphocyte Differentiation. Academic Press. New York. 1976. 17. Makinodan, T. Santos, G. and Quinn, R. Pharm. Rev. 22:189. 1970. 18. Weston, W.L., Claman, N., and Kreuger, G. J. Immunol. 110:880. 1973. 19. Turk, J.L., Parker, D. and Poulter, L. Immunology. 23:493. 1972. 20. Hess, A.D. and Tutsuska, P. J. Immunol. 124:2601. 1980. 21. Larsson, E.L. J. Immunol. 124:2828. 1980. 22. Maugh, T. Science. 210:44. 1980. 23. Kunkel, A. and Klaus, G.G. J. Immunol. 125:2526. 1980. 24. Swain, S. and Dutton, R. J. Immunol. 124:437. 1980. 25. Rich, R.R. and Rich, S.S. J. Immunol. 114:1112. 1975. 26. Rich, S.S. and Rich, R.R. J. Exp. Med. 140:1588. 1974. 27. Sell, S. Ann. Intern. Med. 71:177. 1969. 28. Reinherz, E. and Schlossman, S.F. Cell. 19:821. 1980. 29. Reinherz, E. Rung, P., Goldstein, G., Levey, R. and Schlossman, S.F. PNAS. (USA) 77 : 1588 1980. 30. Katz, D., Graves, M. Dorf, M. Dimenzo, H. and Benacerraf, B. J. Exp. Med. 141:263. 1975. 31. Erb, P. and Feldmann, M. J. Exp. Med. 141:460. 1975. 32. Taussig, M. Isac, M. and Mozes, E. J. Exp. Med. 140:301. 1974. 33. Mozes, E. Isac, R. and Taussig, M. J. Exp. Med. 141:703. 1975. 34. Takemori, T. and Tada, T. J. Exp. Med. 142:1241. 1975.

PAGE 253

237 35. Rich, R. and Pierce C. J. Immunol. 112:1360. 1974. 36. Bach, F., Widmer, M. Bach, ML. and Klein, J. J. Exp. Med. 136:1430. 1972. 37. Klein, J. and Park, J.M. J. Exp. Med. 137:1213. 1973. 38. Ford, W.L. Handbook of Experimental Immunology. Blackwell Scientific. London. 1975. 39. Klein, J. and Schreffler, D. J. Exp. M ed. 135:924. 1972. 40. Demant, P. Folia Biol. (Praha) 16:273. 1970. 41. Tsoi, M.S, Storb, R. Dobbs, S., Meadill, A. and Thomas, E.D. J. Immunol. 125:2258. 1976. 42. Hayry, P. and Def endi V. Science. 168:133. 1970. 43. Hayry, P. Andersson, L.C., Nordling, S. and Virolainen, M. Transplant. Rev. 12:91. 1972. 44. Andersson, L.C. and Hayry, P. Transplant. Rev. 25:121. 1975. 45. Kisielow, P. Nature. 253:219. 1975. 46. Shihu, H., Kisielow, P., Bean, M. Takahashi, T. Boyse, E.A. and Oettgen, H. J. Exp. Med. 141:227. 1975. 47. Swain, S.L., Dennert, G., Wormsley, S. and Dutton, R.W. Eur. J. Immunol. 11:175. 1981. 48. Herzenberg, L.A. and Ledbetter, J. (personal communication to Becton-Dickinson) 1980. 49. Cantor, H. and Boyse, E.A. J. Exp. Med. 141:1376. 1975. 50. Bach, F.H., Bach, M.L., Sondel, P. and Sunderadas, G. Transplant. Rev. 12:30. 1972. 51. Swain, S. and Panfili, p.R. J. Immunol. 122:383. 1979. 52. Hayry, P. Immunogenetics 3:417. 1976. 53. Peck, A.B. and Wigzell, H. Immunogenetics. 6:529. 1978.

PAGE 254

238 54. Peck, A.B., Klein, J. and Wigzell, H. J. Immunol. 125:1978. 1980. 55. Gillis, S. and Smith, K. Nature. 268:154. 1977 56. Baker, P.S., Gillis, S. and Smith, K. J. Exp. Med. 149:273. 1979. 57. Dennert, G. Nature. 277:476. 1979. 58. Glasebrook, A. and Fitch, F. Nature. 278:171. 1979. 59. Glasebrook, A. and Fitch, F. J. Exp. Med. 151:876. 1980. 60. Swain, S.L. and Dutton, R.W. J. Immunol. 124:437. 1980. 61. Watson, J. J. Exp. Med. 150:1510. 1979. 62. Watson, J., Aarden, L., Shaw, J. and Paetkau, V. J. Immunol. 122:1633. 1979. 63. Bonnard, G.D., Yasake, K. and Maca, R. Cell Immunol. 51:390. 1980. 64. Farrar, J., Mizel, S., Fuller-Farrar J., Farrar, W. and Hilfiker, M. J. Immunol. 125:793. 1980. 65. David, T. Yu, Y., Chiorazzi, N. and Kunkel, H. Cell Immunol. 50:305. 1980. 66. Okada, M. and Henney C.S. J. Immunol. 125:850. 1980. 67. Scott, J.W., Ponzio, N., Orosz, C. and Finke, J. J. Immunol 124:2378. 1980. 68. Elliott, B., Kimura, A., Peck, A.B., Nagy, Z. and Wigzell, H. Immunogenetics. 12:525. 1981. 69. Tees, R. and Schreier, M. Nature. 283:780. 1980 70. Taussig, M. Corvalin, J., Binns, R. Roser, B. and Holliman, A. Eur. J. Immunol. 9:768. 1979. 71. Tada, T. and Okumura, K. Adv. Immunol. 28:1. 1979. 72. Taniguiche, M. Takei I. and Tada, T. Nature. 283: 227. 1980.

PAGE 255

239 73. Eschar, Z., Apte, R. Lowry, I., Ben-Nariah, Y. Givol, P. and Mozes, E. Nature. 286:270. 1980. 74. Dennert, G. and Raschke, W. Eur. J. Immunol. 7:352. 1977. 75. von Boehmer, H., Hengartner, H., Nabholty, H., Lernhardt, W. Schreier, M. and Haas. W. Eur. J. Immunol. 9:592. 1979. 76. Sherman, L. J. Exp. Med. 151:1386. 1980. 77. Sredni, B., Tse, H. and Schwartz, R. Nature. 283:581. 1980. 78. Sredni, B., and Schwartz, R. Nature. 287:855. 1980. 79. Fathman, C.G. and Hengartner, H. Nature. 272:617. 1978. 80. Fathman, C.G. and Hengartner, H. PNAS. (USA). 76:5863. 1979. 81. Peck, A.B., Murgita, R. and Wigzell, H. J. Exp. Med. 147:667. 1978. 82. Peck, A.B., Murgita, R. and Wigzell, H. J. Exp. Med. 148:360. 1978. 83. Murgita, R. Goidl, E. Kontainen, S., Beverly, P.C. and Wigzell, H. PNAS. (USA). 75:2897. 1978. 84. Murgita, R. and Tomasi, T. J. Exp. Med. 141:269. 1975. 85. Murgita, R. and Tomasi, T. J. Exp. Med. 141;440. 1975. 86. Lau, L. and Linkins, S. Am. J. Obstet. Gynecol. 124: 533. 1976. 87. Peck, A.B., Murgita, R. and Wigzell, H. J. Immunol. 128:1134. 1982. 88. Skowron-Cendrzak, A. and Ptak, W. Eur. J. Immunol. 6: 451. 1976. 89. Ptak, W. and Skowron-Cendrzak, A. Transplantation 24:45. 1977. 90. Globerson, A. and Umiel, T. Transplantation. 26:438. 1978.

PAGE 256

240 91. Peck, A.B. and Bach, F. J. Inununol Methods. 3:147. 1973. 92. Peck, A.B. and Click, R. Eur. J. Immunol. 3:385. 1973. 93. Peck, A.B. and Wigzell, H. Immunogenetics. 6:529. 1978. 94. Peck, A.B. and Bach, F. Scand. J. Immunol. 4:53. 1975. 95. Shimuzu, S., Konaka, Y. and Smith, R.T. J. Exp. Med. 152:1436. 1981. 96. Rappaport, H., Khalil, A., Halle-Pannenko, 0., Pritchard, L. and Mathe, G. Am. J. Path. 96:121. 1979. 97. Festenstein, H. Transplant Rev 3:74. 1970. 98. Festenstein, H. Transplant Rev 15:62. 1973. 99. Festenstein, H. Transplant Proc 8:339. 1970. 100. Molnar-Kimber, K. and Sprent, J. J. Exp. Med. 15:407. 1981. 101. Molnar-Kimber, K. and Sprent, J. Transplantion. 31:376. 1981. 102. Nakayana, E., Shiku, H., Stockert, E. Oettgen, H.F. and Old, L.J. PNAS. (USA). 76:1977. 1979. 103. Shinohara, N. and Sachs, D.H. J. Exp. Med. 150:432. 1979. 104. Hard, R.C. Transplantation. 17:613. 1974. 105. Peck, A.B., Smith, R.T. and Jadus M.R. (submitted for publication) J. Immunol 1982. 106. Peeler, K.L., Wigzell, H. and Peck, A.B. Scand. J. Immunol, (in press). 1982. 107. Wagner, H., Hardt, C, Heeg, K., Pfizenmaier, K., Stotter, H. and Rollinghoff, M. Immunobiol 161:139. 1982.

PAGE 257

241 108. Hardt, C, Rollinghoff, M. PFizenmaier K., Mosman, H. and Wagner, H. J. Exp. Med 154:262. 1981. 109. Bach, F.H., Bock, H., Graupner, K., Day, E and Klosterman, H. PNAS. (USA) 62:377. 1969. 110. Wolters, E.A.J, and Banner, R. Transplantation. 26:40. 1978. 111. Wolters, E.A.J, and Benner, R. Nature. 279:642. 1979. "• 112. Ford, W.L. and Gowans, J.L. Sem. Haematol. 6:67. 1973. 113. Brondz, B.D., Andreev, A.V., Egorova, S.G. and Drizlikh, G.I. Scand. J. Immunol. 10:195. 1979. 114. Cerrotini, J.C. and Brunner, K.T. Adv. Immunol. 18:67. 1974. 115. Hamilton, B.L., Bevan, M.J. and ParJcman, R. J. Immunol. 126:621. 1981. 116. Hamilton, B.L. and Parkman, R. J Immunol 128:376. 1982. 117. Henney, C.S., Kuribayashi, K., Kern, D.E. and Gillis, S. Nature. 291:335. 1981. 118. Gillis, S., Smith, K.A. and Watson, J. J. Immunol. 124: 1254. 1980. 119. Watson, J., Gillis, J., Marbrook, D., Mochizuki, D. and Smith, K.A. J. Exp. Med. 150:849. 1979. 120. Nabel, G., Galli, S.J., Dvorak, H.F. and Cantor, H. Nature. 291:332. 1981. 121. Delovitch, T.L., Harris, J.F., Battistella, R. and Kaufman, K. J. Exp. Med. 155:61. 1982. 122. Elkins, W.L. Transplant. Proc. 8:343. 1976. 123. Levy, R.B. and Hansen, T.H. Immunogenetics. 10:7. 1982. 124. Loveland, B.E. and McKenzie, I.F.C. Transplantaion. 33: 217. 1982.

PAGE 258

242 125. Loveland, B.E., Hogarth, P.M., Ceredig, R. and McKenzie, I.F.C. J. Exp. Med. 153:1044. 1981. • 126. Mason, D.W. Transplantation. 32:222. 1981. 127. Dallman, M.J. and Mason, D.W. Transplantation 33:221. 1982. 128. de Sousa, M. Lymphocyte Circulation. Academic Press. New York. 1982. 129. Ihle, J., Rebar, L., Keller, J., Lee, J.C and Hapel, A. Immunol. Rev. 63:5. 1982. 130. Garland, J.M. Immunology Today. 3:208. 1982. 131. Fauci, A.S. Am. J. Med. 64:729. 1978. 132. Sprent, J. and Miller, J.F.A.P. Cell. Immunol. 3:361. 1972. 133. Sprent, J. and Miller, J.F.A.P. Cell. Immunol. 3:385. 1972. 134. Sprent, J. and Miller, J.F.A.P. Cell. Immunol. 3:213. 1972. 135. Sprent, J. and Miller, J.F.A.P. J. Exp. Med. 139:1. 1974. 136. Rybcznska, Z., Sajak, A. and Skowron-Cendrzak, A. Arch. Immunol. Et Therap. Experiment. 29:341. 1981. 137. Mosier, D.E. and Johnson, B.M. J. Exp. Med. 141:216. 1975. 138. Basset, T.M., Coons, T.A. Goldberg, E.H. and Williams, J. J. Immunol. 119:1855. 1977. 139. Rodrigeuz, G., Andersson, G., Wigzell, H. and Peck, A.B. Eur. J. Immunol. 9:737, 1979. 140. Pavia, C.S. and Stites, D.P. Cell. Immunol. 42:48. 1978. 141. Arygris, B. Cell. Immunol. 62:412. 1981 142. Arygris, B. Cell. Immunol. 66:352. 1982 143. Arygris, B. Transplantation. 31:334. 1981

PAGE 259

243 144. Matossian-Rogers, A. and Festenstein, H. J. Exp. Med. 143:456. 1976. 145. Arygris, B. Cell. Immunol. 57:62. 1981. 146. Click, R.E. Tranplant. Proc. 11:490. 1979. 147. Sprent, J. J. Exp. Med. 148:478. 1978. 148. Korngold, R. and Sprent, J. J. Exp. Med 151:1114. 1980. 149. Sprent, J. and Korngold, R. Immuno logy Today. 2:189. 1981.

PAGE 260

BIOGRAPHICAL SKETCH Martin Jadus was born on January 10, 1953 in Girardville, Pennsylvania. He lived in Wilmington, Delaware until he graduated from the University of Delaware with Bachelor of Science Degrees in Biological Sciences and Chemistry in 1976. From September 1976 to August 1978 he lived in Melbourne Florida and received a Master of Science Degree in the Biological Sciences from Florida Institute of Technology. From 1978 until 1983 he attended the University of Florida. At present he plans to either move to the West Coast or to Uppsala, Sweden in order to continue active research in Immunology. 244

PAGE 261

I certify that I have read this study and that in my opinion it confirms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. /\utMo\\ O X^iz. Ammon B. Peck, Chairman Assistant Professor of Pathology I certify that I have read this study and that in my opinion it confirms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. C2^ a.iA^ Paul A. Klein, Associate Professor of Pathology I certify that I have read this study and that in my opinion it confirms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Edward K. Wakeland, Assistant Professor of Pathology

PAGE 262

I certify that I have read this study and that in my opinion it confirms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Noel K. Maclaren, Professor of Pathology. I certify that I have read this study and that in my opinion it confirms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Roy S. Weiner, Professor of Immunology and Medical Microbiology

PAGE 263

This dissertation was submitted to the Graduate Faculty of the College of Medicine and to the Graduate Council, and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. August, 1983 Dean, College of Medicine >^C^?Yfi~
PAGE 264

UNIVERSITY OF FLORIDA 3 1262 08554 7858


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EBLXABU0F_UUBVXV INGEST_TIME 2015-03-27T18:36:59Z PACKAGE AA00029773_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES