Citation
Comparison of Extracellular Matrix Proteins from Allyl Alcohol and Carbon Tetrachloride

Material Information

Title:
Comparison of Extracellular Matrix Proteins from Allyl Alcohol and Carbon Tetrachloride
Creator:
Brown, Alicia Renae
Place of Publication:
[Gainesville, Fla.]
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (107 p.)

Thesis/Dissertation Information

Degree:
Master's ( M.S.)
Degree Grantor:
University of Florida
Degree Disciplines:
Medical Sciences
Medicine
Committee Chair:
Petersen, Bryon E.
Committee Members:
Terada, Naohiro
Peck, Ammon B.
Graduation Date:
5/1/2008

Subjects

Subjects / Keywords:
Cell growth ( jstor )
Cells ( jstor )
Collagens ( jstor )
Connective tissues ( jstor )
Ellipses ( jstor )
Fibrosis ( jstor )
Hepatocytes ( jstor )
Liver ( jstor )
Physical trauma ( jstor )
Rats ( jstor )
Medicine -- Dissertations, Academic -- UF
allyl, tetrachloride
Genre:
Electronic Thesis or Dissertation
bibliography ( marcgt )
theses ( marcgt )
Medical Sciences thesis, M.S.

Notes

Abstract:
Allyl alcohol (AA) and carbon tetrachloride (CCl4) are two chronic liver injury models that cause extensive periportal and centrilobular damage, respectfully. The extracellular matrix (ECM) is a complex structure aiding in cell activation, migration and differentiation within the liver. The composition of this matrix varies throughout the liver, corresponding to cellular and functional requirements within each region of the organ. During the liver regeneration process, the ECM goes through substantial changes which can provide evidence to the signals required for restoration of the liver mass. Transforming growth factor-beta and connective tissue growth factor have been shown to be key cytokines in the regulation of this process. Cells attach to the ECM by means of transmembrane glycoproteins called integrins. The extracellular portion of integrins binds to various types of ECM proteins including collagens, laminins and fibronectin. The current study examines the role of molecular signals and ECM components in each of these two chronic injury models. Archival, formalin fixed paraffin embedded liver tissue from AA and CCl4 treated rats was examined. Attempts at nucleic acid and protein extraction from these tissues were unsuccessful. Therefore, immunohistochemistry was used to describe the fibrotic response. AA and CCl4 proved to be very different models of chronic liver disease, as one resulted in massive necrosis (AA) while the other resulted in cirrhosis (CCl4) as evidenced of Trichrome staining. The AA model developed a lose ECM at the site of injury that was devoid of collagen type IV, where as CCl4 developed a dense fibrotic scar that was rich in collagen type IV. Understanding the composition of ECM during chronic liver injury could lead to better methods for the treatment of pathologies involving hepatic fibrosis. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (M.S.)--University of Florida, 2008.
Local:
Adviser: Petersen, Bryon E.
Statement of Responsibility:
by Alicia Renae Brown

Record Information

Source Institution:
UFRGP
Rights Management:
Copyright Alicia Renae Brown. 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.
Embargo Date:
7/11/2008
Classification:
LD1780 2008 ( lcc )

Downloads

This item has the following downloads:


Full Text





Trichrome staining. The AA model developed a lose ECM at the site of injury that was devoid of

collagen type IV, where as CCl4 developed a dense fibrotic scar that was rich in collagen type

IV. Understanding the composition of ECM during chronic liver injury could lead to better

methods for the treatment of pathologies involving hepatic fibrosis.









liver regeneration [70]. FN has been shown to bind IGFBP3 and to form ternary complexes with

IGF-1 in human plasma to possibly sequester them into tissues [71].

Laminin

Laminin (LM) is a large ECM glycoprotein found in basement membranes of epithelia,

surrounding blood vessels and nerves in established tissues. This protein interacts with cell

surface receptors and has roles in cell migration during embryonic development and tissue

organization [72]. LM subunits (A, Bl, B2, S and M) in the perisinusoidal space of the rat liver

have been detected in small streaks of basement membranes extending from the portobiliary tract

and to a lesser degree from the central vein and perisinusoidal spaces [72]. LM is co-expressed

with CollV. LM is also a prominent component of the perisinusoidal matrix during development,

injury and regeneration of the liver. It is present within the cytoplasm of hepatocytes, SEC and

HSC during these times [19]. In fibrosis, the HSC are the main contributors oflaminin. Due to its

many isoforms of each subunit, it is difficult to view every possible place it is located.

LMs are secreted and incorporated into cell-associated extracellular matrices. These

proteins form independent networks associated with CollV via nidogen [73] and through light,

direct interactions with each other [74]. LMs also bind to cell membranes through integrin

receptors and other plasma membrane molecules, such as the dystroglycan glycoprotein complex

and Lutheran blood group glycoprotein [75]. Through these interactions, laminins critically

contribute to cell attachment and differentiation, cell shape and movement, maintenance of tissue

phenotype, and promotion of tissue survival. It has been shown in vitro that hepatocytes have an

increased affinity for laminin post PH [76]. The main role of laminin may be to provide cell

attachment for proper tissue organization during liver regeneration [77; 75]. Work done by

Shakado in 1995 showed that laminin was crucial for SECs to form long tubular structures while

advancing into a gel matrix [75].







































To my parents, Gary and Julie Brown.











Activated Hepatic Stellate Cells

m Allyl Alcohol
mCarbon Tetrachlorlde
120-
100
S80
60
E 40

S,,

dl5 d45 d90
Days of Tre atment


Figure 3-9 Activated HSCs over the time course of chronic AA (blue) and CCl4 (pink) liver
injuries. Figure 3-10: Trichrome staining for Collagen I during chronic liver injury.
(Al-3) AA injury at dl5, 45 and 90, respectively. (Bl1-3) CCl4 injury at dl5, 45 and
90, respectively. (C 1-2) NRL. (D-E) Higher magnification of d90 AA and D90 CCl4,
respectively. A and B original magnification=600X, Cl original magnification=200X,
C2-F original magnification=600X.



























Figure 3-2 Hematoxylin and Eosin analysis of AA (A) and CCl4 (B) treated rat livers 90 days
following initiation of injury. Time course of chronic AA injury: Al=dl5, A2=d30,
A3= d45, A4= d60, A5= d70, A6= d90. Time course of chronic CCl4 injury: B l= dl5,
B2=d30, B3=d45, B4=d60, B5=d70, B6=d90. Labeling points out the central vein
(CV) and portal triad (PT). A and B, original magnification=200X. Al-B6, original
magnifi cati on= 100X.









162. Ujike K, Shinji T, Hirasaki S, Shiraha H, Nakamura M, Tsuji T, Koide N. Kinetics of
expression of connective tissue growth factor gene during liver regeneration after partial
hepatectomy and D-galactosamine-induced liver injury in rats. Biochem Biophys Res
Commun. 2000;277(2):448-54.

163. Paradis V, Dargere D, Bonvoust F, et al. Effects and regulation of connective tissue
growth factor on hepatic stellate cells. Lab Invest 2002;82:767-774.

164. Hayashi N, Kakimuma T, Soma Y, Grotendorst GR, Tamaki K, Harada M, and Igarashi
A. Connective tissue growth factor is directly related to liver fibrosis.
Hepatogastroenterology 2002;49:133-135.

165. Crean JK, Finlay D, Murphy M, Moss C, Godson C, Martin F, Brady HR. The role of
p42/44 MAPK and protein kinase B in connective tissue growth factor induced
extracellular matrix protein production, cell migration, and actin cytoskeletal
rearrangement in human mesangial cells. J Biol Chem. 2002;277(46):44187-94.

166. Segarini PR, Nesbitt JE, Li D, Hays LG, Yates JR 3rd, Carmichael DF. The low density
lipoprotein receptor-related protein/alpha2-macrogl obulin receptor i s a receptor for
connective tissue growth factor. J Biol Chem. 2001;,276(44):40659-67.

167. Shupe T, Sell S. Low hepatic glutathione S-transferase and increased hepatic DNA
adduction contribute to increased tumorigenicity of aflatoxin B1 in newborn and partially
hepatectomized mice. Toxicol Lett. 2004 14; 148(1-2): 1-9.

168. Kinnman N, Housset C. Peribiliary myofibroblasts in biliary type liver fibrosis. Front
Biosci. 2002;7:496-503.









LIST OF REFERENCES


1. Hepatology: A Textbook of Liver Disease. Boyer and Zakim (eds), Saunders. New York
(2003).

2. Crawford JM, Robbins and Cotran. Pathologic Basis of Disease. Kumar,V., Abbas,A. &
Fausto N. (eds.), pp. 877-938, Elsevier Saunders, Philadelphia (2005).

3. Rappaport AM. Acinar units and the pathophysiology of the liver. In: The liver
morphology, biochemistry, physiology. Vol 1. Rouiller CH. (eds.), pp. 265-328,
Academic Press, New York. (1963).

4. Saxena R, Theise ND, Crawford JM. Microanatomy of the human liver-exploring the
hidden interfaces. Hepatology. 1999;30(6):1339-46.

5. Fausto N. The Liver biology and pathobiology. Arias,I.M. and Boyer,J.L. (eds.), pp.
1501-1518, Raven Press, New York (1994).

6. Overturf K, al-Dhalimy M, Ou CN, Finegold M, Grompe M. Serial transplantation
reveals the stem-cell-like regenerative potential of adult mouse hepatocytes. Am J Pathol.
1997;151(5):1273-80.

7. Tateno C, Yoshizato K. Growth potential and differentiation capacity of adult rat
hepatocytes in vitro. Wound Repair Regen. 1999;7(1):36-44.

8. Ogi M, Yokomori, Oda M, Yoshimura K, Nomura M, Ohshima S, Akita M, Toda K, Ishii
H. Distribution and localization of caveolin-1 in sinusoidal cells in rat liver. Med
Electron Microsc. 2003;36(1):33-40.

9. Kurose I. Miura S, Saito H, et al. Rat Kupffer cell-derived nitric oxide modulates
induction of lymphokine-activated killer cell. Gastroenterology 1995;109: 1958-1968.

10. Paul SC, Lv P, Xiao YJ, An P, Liu SQ, Luo HS. Thalidomide in rat liver cirrhosis:
blockade of tumor necrosis factor-alpha via inhibition of degradation of an inhibitor of
nuclear factor-kappaB. Pathobiology. 2006;73(2):82-92.

11. Edwards MJ, Keller BJ, Kauffman FC, and Thurman RG. The involvement of Kupffer
cells in carbon tetrachloride toxicity. Toxicol Appl Pharmacol 1993;119:275-279.

12. Winau F, Hegasy G, Weiskirchen R, Weber S, Cassan C, Sieling PA, Modlin RL, Liblau
RS, Gressner AM, Kaufmann SH. Ito cells are liver-resident antigen-presenting cells for
activating T cell responses. Immunity 2007;26(1): 117-29.

13. Roth S, Michel K, Gressner AM. (Latent) transforming growth factor-beta in liver
parenchymal cells, its injury-dependent release and paracrine effects on hepatic stellate
cells. Hepatology 1998;27:1003-12.










TGFP, CTGF expression was seen to consistently increase throughout the time course of the

study. While CTGF has classically been considered to be produced mainly by hepatic stellate

cells, recent studies confirm that hepatocytes may express this factor in response to hepatic

injury [88]. In our injury models, hepatocytes do, indeed, appear to be the maj or source of CTGF

(Figure 3-7).

TGFP and CTGF are critical in the activation of hepatic stellate cells to a matrix

synthesizing phenotype. A hallmark of hepatic stellate cell activation is the expression of a-

smooth muscle actin (SMA). In the CCl4 mOdel, a-SMA' cells are easily identified, even in the

earliest time points (Figure 3-8 Bl-3). The number of these cells generally increases over the

course of injury. a-SMA' cells are seen throughout the fibrotic regions of the liver particularly

within the fibrous cords that encircle the regenerative nodules. It is worth noting that a-SMA+

hepatic stellate cells are found infiltrating these nodules in the later time points (Figure 3-8 F). It

is possible that these stellate cells provide matrix for organization of the hepatic micro-

architecture within the nodules. In contrast to the CCl4 mOdel, the AA model shows very few a-

SMA+ stellate cells in the early time points. While the number of these cells does increase over

the time course of injury, it never reaches the numbers seen in the CCl4 mOdel (Figure 3-8 Al-3,

Figure 3-9). The stellate cells are the primary source of ECM seen in these models; it is not

surprising that fewer stellate cells are present in the less fibrotic AA model. It is also possible

that the a-SMA+ fibroblasts in the AA model are not of stellate cell origin but are myofibroblasts

that surround bile ducts and vessels in the portal triad [168] and may not perform in the same

manner.

Hepatic stellate cell activation ultimately leads to matrix deposition. The maj or

constituents of this matrix are collagens, laminins and fibronectin. The pathologic accumulation









ferrous sulfate, high levels of exposure from yellow phosphorus, acetic acid and aspirin, and

acute viral hepatitis associated with cocaine and labetalol [94].

AA oxidation occurs through alcohol dehydrogenase (ADH). The periportal region

possesses higher concentrations of ADH which correlates with the periportal necrosis seen by

AA exposure. AA exerts a dose-dependent toxicity on the cells which is inversely related to

cellular glutathione (GSH) [95]. GSH is a ubiquitous tripeptide (including a cysteine) that acts as

a powerful antioxidant. It exists in almost every cell of the body, and without it other

antioxidants like vitamin E and C are not able to function efficiently [19]. AA toxicity can be

prevented by inhibitors of ADH and augmented by the aldehyde dehydrogenase (ALDH)

inhibitor disulfiram, which lead to the discovery that toxicity is produced by a metabolite,

specifically the aldehyde acrolein (CH2=CHCH=0) as noted from many references [96; 91; 97].

Acrolein is oxidized by ALDH and has been shown to require NAD' [98]. This yields free

radicals, and protons are transferred to selenium oxide or alpha carbonyl compounds like ketones

which are converted to diketones. ALDH are grouped into classes (ALDH1-7), and there are

around 17 ALDH enzymes in human [99]. ALDH2, 3 and 5 are capable of detoxifying acrolein

[100]. Since ALDH2 in rat liver is more concentrated in periportal than in perivenous cells [101]

and NAD 'is at a higher concentration around the periportal regions, this region is targeted and

damaged. ALDH2 has a low km for acetaldehyde and catalyzes most acetaldehyde oxidation in

the liver. It is expressed constitutively, is present exclusively in mitochondria and is NAD -

dependent [93]. Acrolein, with ALDH, is an excellent substrate for glutathione-S-transferases

[92] and forms adducts with GSH and protein thiols in isolated hepatocytes [95]. In cells

depleted of GSH, acrolein may react with essential macromolecules and thereby lead to

structural and functional derangement and, eventually, irreversible injury. Binding of acrolein to









Hepatocyte and Non-Parenchymal Cell Proliferation in Chronic AA and CCl4 Models

Cell proliferation during the time course of AA and CCl4 chronic injury was determined

by BrdU incorporation into newly synthesized DNA. BrdU labeling of NRL and rat liver 24hr

following PH were used as negative and positive controls, respectively. Analysis of the BrdU

staining pattern in AA treated livers indicates concentrations of proliferating cells within

localized pockets, highlighting the need to count many different microscopic fields. This is best

demonstrated in the non-parenchymal fraction of AA d90 (Figure 3-3) where several fields

contained greater than 200 proliferating cells while the average was only 91. In the CCl4 mOdel,

two waves of hepatocyte proliferation were noted with peaks at d45 and d90 (Table 3-1). The

AA model also involved an ebb and flow with respect to hepatocyte proliferation, though this

was much less pronounced than in the CCl4 mOdel. Non-parenchymal cell proliferation appears

to lag behind hepatocyte proliferation in these models with a robust peak of proliferation noted at

d90 in the AA model. As expected, normal rat liver shows very little proliferation (Figure 3-4C).

Partial hepatectomy rat liver, 24hr following surgery demonstrates massive proliferation of the

hepatocyte cell fraction (Figure 3-4D). A peak of non-parenchymal cell proliferation is noted

several days later (data not shown).

Alpha Fetoprotein Expression by Hepatic Oval Cells in Chronic AA and CCl4 MOdels

To determine the degree of oval cell participation in these two injury models, the oval

slides were stained for the oval cell marker AFP. Adult NRL contains very few, if any, AFP

cells, and was used as a negative control (Figure 3-5C). The AA model has previously been

shown to induce an oval cell response. In agreement with this notion, a population of AFP

hepatic oval cells develops over the time course of chronic AA injury (Figure 3-5A). Analysis of

higher magnification fields shows the AFP' oval cells in the AA model are arranges in duct like

structures (Figure 3-5E). These structures are also found in the CCl4 mOdel, though they are quite
















































Figure 3 11 Collagen type IV in chronic liver inj ury models by immunohi stochemi stry. (Al1-3)
AA injury at dl5, 45 and 90, respectively. (Bl-3) CCl4 injury at dl5, 45 and 90,
respectively. (C 1-2) NRL. (D) immunoglobulin control. (E-F) Higher magnification
of d90 AA and D90 CCl4, respectively. A and B original magnification=600X, Cl
original magnification=200X, C2-F original magnification=600X.









concentrations of .25-1CIL, dNTPs at 0.6C1L, 2C1L Eppendorph's smart taq buffer with magnesium

balance, 0.2C1l Fisher brand Taq in a 20C1L reaction. GAPDH primers were as followed: Forward:

ATCACTGCCACTCAGAAGAC ; Reverse: AACACTGAGCATC TC CCTCA. The PCR was

performed at a melting temperature of 95oC for 5min followed by 25-40 cycles of 95oC for

30sec, 580C for 1min and 72oC for 1-1.5min and Einished by 10min at 72oC. The samples were

cooled to RT or stored in 4oC before being loaded into a 1.5-2% agarose gel with 1% ethidium

bromide.

Histological Analysis

Hematoxylin and Eosin Staining of Paraffin Embedded Tissue

Tissue sections of 5CLM in size were cut and placed in a 42oC water bath. There were then

lifted from the bath with a SuperfrostTM Plus (Thermo Fisher Scientifie Inc., Waltham, MA)

positively charged slide. The slides were air dried overnight at RT. The paraffin was removed

and the slides rehydrated by incubating them in Xylene 2 x 5min, 100% ethanol 2 x 2min, 95% x

ethanol 2 x Imin, and distilled H20 for 1min. If an antigen retrieval step is necessary, it is

performed at this point then rinsed with distilled water before proceeding. Nucleic acids and

other positively charged molecules were then stained with Hematoxylin 7211 (Richard-Allan

Scientific, Kalamazoo, MI) for 2min 15sec and rinsed with distilled H20 for 2 x Imin. The blue

color of the Hematoxylin was intensified by incubating the slides in ClarifierTM 1 (Richard-Allan

Scientific) for 1min, distilled H20 for 1min, Bluing Reagent (Richard-Allan Scientific) for 1min,

distilled H20 for 1min, and 80% EtOH for 1min. Proteins were then stained a pink color with

Eosin-Y (Richard-Allan Scientific) for 1min 30sec. The tissue was then dehydrated for

coverslipping with 2 x Imin 95% EtOH, 2 x Imin 100% EtOH, and 3 x Imin Xylene. Coverslips

were then applied with CytosealTM XYL (Richard-Allan Scientific).









involving p38, and focal adhesion kinase signaling cascades. Together, these regulate the

proliferative response, activating cell cycle progression as well as collagen gene expression.

SMAD and p38 MAPK signaling have been found to independently and additively regulate

collagen type I gene expression by transcriptional activation while p38 MAPK, but not SMAD

signaling, increases collagen mRNA stability, leading to increased synthesis and deposition [87].

Fibroblasts other than HSCs are involved in hepatic fibrosis. The possibility of epithelial-

mesenchymal transition (EMT), which describes the transition of biliary epithelial cells or even

hepatocytes to fibroblasts, is still under debate. The fibroblasts from EMT actively participate in

the generation of fibrotic ECM as established in lung and kidney fibrosis [88]. The role of EMT

in hepatic fibrosis remains unknown.

The ECM is a complex structure aiding in the support, maintenance and regeneration of the

liver. This matrix varies throughout the hepatic architecture to correspond to vast functions of

this organ. The distribution of the ECM and its associated proteins is important to understand

how the liver responds to injury. TGFP and CTGF are the key factors for the j ob. AA and CCl4

are two chronic injury models that demonstrate this occurrence [89; 90]. These two injury

models cause damage to different regions of the liver. Each chemical is metabolized by utilizing

molecules specific for their classification. However they both begin with an oxidation step.

Chronic Allyl Alcohol Exposure

AA is the smallest representative of allylic alcohols (CH2=CHCH20H). This chemical is

used in industries as a synthetic intermediate, an effective herbicide and pesticide. AA

intoxication leads to periportal (piecemeal) necrosis, ductular proliferation, abundant connective

tissue development, macro nodules, cirrhosis, calcium influxes and minimal oval cell inHiltration

[1; 19; 91; 92; 93]. This is a very useful model for periportal injury because few chemicals cause

this type of damage. Some examples of periportal injuries come from iron overloading from









26. Saperstein LA, Jirtle RL, Farouk M, Thompson H J, Chung KS, Meyers W C.
Transforming growth factor-b 1 and mannose 6-phosphate/insulin-like growth factor-II
receptor expression during intrahepatic bile duct hyperplasia and biliary fibrosis in the
rat. Hepatology 1994; 19:412-417.

27. Sedlaczek N, Jia JD, Bauer M, Herbst H, Ruehl M, Hahn EG, Schuppan D. Proliferating
bile duct epithelial cells are a maj or source of connective tissue growth factor in rat
biliary fibrosis. Am J Pathol. 2001;1 58(4):123 9-44.

28. Michalopoulos GK, Barua L, Bowen WC. Transdifferentiation of rat hepatocytes into
biliary cells after bile duct ligation and toxic biliary injury. Hepatology 2005;41(3):535-
44.

29. Farber E. Similarities in the sequence of early histologic changes induced in the liver of
the rate by ethi onine, 2AAF, and 3 -methyl -4-dimethyl aminoazb enzene. Cancer Res.
1956;16:142.

30. Michalopoulos GK, DeFrances MC. Liver regeneration. Science. 1 997;276(5309):60-6.

31. Oh SH, Muzzonigro TM, Bae SH, LaPlante JM, Hatch HM, Petersen BE. Adult bone
marrow-derived cells trans-differentiating into insulin-producing cells for the treatment
of type I diabetes. Lab Invest. 2004;84(5):607-17.

32. Hixson DC, Allison JP. Monoclonal antibodies recognizing oval cells induced in the liver
of rats by N-2-fluorenylacetamide or ethionine in a choline-deficient diet. Cancer Res
1985;45:3750.

33. Petersen BE, Zaj ac FV, Michalopoulos GK. Hepatic oval cell activation in response to
injury following chemically induced periportal or pericentral damage in rats. Hepatology
1998;27:1030.

34. Oh SH, Hatch HM, Petersen BE. Hepatic oval 'stem' cell in liver regeneration. Seminars
in CDB. 2002;13:405-409.

35. Petersen BE, Zaj ac VF, Michalopoulos GK. Extracellular matrix remodeling at the early
stages of liver regeneration in the rat. Am J Pathol. 1997; 151:905.

36. Shinozuka H, Lombardi B, Sell S, lammarino RM. Early histological and functional
alterations of ethionine liver carcinogenesis in rats fed a choline-deficient diet. Cancer
Res. 1978;38(4):1092-8.

37. Hixson DC, Chapman L, McBride A, Faris R, Yang L. Antigenic phenotypes common to
rat oval cells, primary hepatocellular carcinomas and developing bile ducts.
Carcinogenesis. 1997;18(6): 1169-75.









motif (GCGCCXXC) that is involved in binding insulin like growth factor (IGF) with low

affinity [144]. The physiological role of this binding site still remains to be defined. Module 2

contains a von Willebrand type C domain (vWC) that occurs in von Willebrand factor (vWF) as

well as various mucins, thrombospondins, and collagens [143] and may be involved in

oligomerization [145]. Module 3 is a thrombospondin type 1 domain (TSPl) that contains the

local motif WSXCSXXCG and appears to be a cell attachment factor that binds sulfated

glycoconjugates [143; 146]. Module 4 is a C-terminal module that also occurs in the C-termini of

a variety of unrelated soluble proteins including TGFP and PDGF [143] and is responsible for

dimerization and receptor-binding [143]. The bioactive form of porcine CTGF is contained

within the 103 C-terminal residues of the primary translational product [147] and thus supports

the proposed role of module 4 in binding cell surface receptors [143]. Pi et al reported that the C-

and N-terminal ends of CTGF are able to bind fibronectin [148].

Biological Significance of Connective Tissue Growth Factor

CTGF is associated with immediate early growth responsive genes that putatively regulate

the proliferati on/differenti ati on of vari ous connective ti ssue cell types during embryogenesi s

[147]. It is a cysteine-rich, matrix-associated, heparin, perlacan, fibronectin and integrin-binding

protein that promotes endothelial cell growth, migration, adhesion and survival, while

stimulating ECM production [149], chemotaxis, proliferation and angiogenesis. However, CTGF

can act as cell growth inhibitor and induce apoptosis. In some cases, it is intrinsically

nonmitogenic and augments the activity of other growth factors [150]. In addition, CTGF was

shown tobe positively regulated by vascular endothelial growth factor, epidermal growth factor,

fibroblast growth factor [151], plasma clotting factor VIIa, thrombin [152; 153] and by

lysophosphatidic acid and serotonin activation of heptahelical receptors [154] but negatively

regulated by tumor necrosis factor-a [155].












TABLE OF CONTENTS


page

ACKNOWLEDGMENTS .............. ...............4.....


LIST OF TABLES ................. ...............8.__. .....


LIST OF FIGURES .............. ...............9.....


LI ST OF AB BREVIAT IONS .............. ................. 10...._......


AB S TRAC T ............._. .......... ..............._ 12...


CHAPTER


1 LITERATURE REVIEW ................. ...............14...............


Anatomy and Physiology of the Liver ..........._...__.......... ...............14...
Cell s of the Liver ......_._ ................ ...............16. ....

Hepatocytes .............. ...............16....
Endothelial Cells .............. ...............17....
Immune Cell s............... ...............17.
Stellate Cells............... ...............18.
Biliary Cell s............... ...............20.
Oval Cell s .............. ...............2 1....
Liver Regeneration .............. ....... ...............23
Overview of the Extracellular Matrix ........._... ...... ._._ ...............27..

Alpha- Smooth Muscle Actin ........._.._ ..... ._._ ...............28...
Proteins of the Extracellular Matrix ........._.... .....___ .....__. ..... ... .............29
Collagens ............. ...... __ ...............29....
Collagen Type IV .............. ...............29....
Fibronectin............... ..............3
Lam inin ............... ..... .... ............... 1....
Basement Membranes in the Liver ............. ...... ._ ...............32...
Introduction to Liver Fibrosis ............. ...... ._ ...............32...
Chronic Allyl Alcohol Exposure .............. ...............35....
Chronic Carbon Tetrachloride Exposure ................. ...............37................
Degradation of Pre-Existing Matrix ................. ............ ...............39......
Overview of the Transforming Growth Factor Beta ................. .............. ......... .....40
Activation of Transforming Growth Factor Beta .............. ...............41....
Biological Activity of Transforming Growth Factor Beta .............. ...............42....
Overview of Connective Tissue Growth Factor ................. .......... ............... 43. ...
Biological Significance of Connective Tissue Growth Factor ................. ............ .........44
Specific Aims of this Study ....._.. ................ ...............46.....

2 MATERIALS AND METHODS .............. ...............48....









within the portal zone indicating a relatively diffuse ECM (Figure 3-10 D). IHC for collagen type

IV shows a similar staining pattern in the AA injury model (Figure 3-11 Al-3). Interstitial

collagen staining in the CCl4 mOdel demonstrates progressive deposition of this ECM component

throughout the time course of injury (Figure 3-10 Bl-3). Staining appears to be much more

intense in this model indicating a well organized, dense accumulation of ECM surrounding and

bridging the central zones (Figure 3-10 E). The staining pattern for collagen IV is very similar to

interstitial collagen with progressive pericentral deposition and concentration at the later time

points (Figure 3-1 1 Bl-3, F). Accumulations of collagen, clearly border the regenerative nodules

in the CCl4 mOdel, where as a very light deposition of collagen is present within the necrotic

zones of the AA model (Figure 3-10 D, E). NRL displayed the expected staining patterns for

these collagens with strong concentrations found bordering vessels and within the basement

membrane of the biliary tree (Figures 3-10, 3-11 C1-2).

Laminin Expression by Hepatic Cells in Chronic AA and CCl4 MOdels

Laminin is another component of ECM. This matrix protein has been shown to play a role

in migration of cells to regions of injury. Laminin staining was used as an indication of cell

trafficking potential through fibrotic regions. NRL was used as a reference for comparison

(Figure 3-12 C1-2). In the AA model, laminin does not appear to accumulate throughout the time

course of injury (Figure 3-12 Al-3). Diffuse laminin staining within the periportal necrotic zone

is likely non-specific (Figure 3-12 A3, E). Conversely, strong progressive laminin deposition is

apparent in the pericentral fibrotic regions of the CCl4 mOdel (Figure 3-12 Bl-3, F). Laminin has

also been used to identify the organization of hepatic parenchyma. Ordered cords of hepatocytes

are observed in NRL (Figure 3-12 C2). It is interesting to note that hepatocytes within the

regenerative nodules at later time points of the CCl4 injury do not appear to be well organized

(Figure 3-12 F).











ACKNOWLEDGMENTS

I express my most sincere gratitude to my mentor Dr. Bryon Petersen for his great

perspectives on life and science. I also thank my committee members for all of the insights on

my proj ect. The post doctorates in my lab, (Drs. Seh-Hoon Oh, Liya Pi, Anna Piscaglia, Thomas

Shupe, and Jennifer Williams) have been a tremendous help to my success. My peers (Houda

Darwiche, Susan Ellor, Dana Pintilie, and Nicole Steiger) gave me experimental advice,

encouragement, and friendship over the past 2 years. I would also like to thank my parents, Gary

and Julie Brown for always supporting everything I have done, and my sisters Christina and

Andrea for their consistent support.









Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

COMPARISON OF EXTRA CELLULAR MATRIX PROTEINS FROM ALLYL ALCOHOL
AND CARBON TETRACHLORIDE

By

Alicia Renae Brown

May 2008

Chair: Bryon Petersen
Major: Medical Sciences

Allyl alcohol (AA) and carbon tetrachloride (CCl4) are two chronic liver injury models

that cause extensive periportal and centrilobular damage, respectfully. The extracellular matrix

(ECM) is a complex structure aiding in cell activation, migration and differentiation within the

liver. The composition of this matrix varies throughout the liver, corresponding to cellular and

functional requirements within each region of the organ. During the liver regeneration process,

the ECM goes through substantial changes which can provide evidence to the signals required

for restoration of the liver mass. Transforming growth factor-beta and connective tissue growth

factor have been shown to be key cytokines in the regulation of this process. Cells attach to the

ECM by means of transmembrane glycoproteins called integrins. The extracellular portion of

integrins binds to various types of ECM proteins including collagens, laminins and fibronectin.

The current study examines the role of molecular signals and ECM components in each of these

two chronic injury models. Archival, formalin Eixed paraffin embedded liver tissue from AA and

CCl4 treated rats was examined. Attempts at nucleic acid and protein extraction from these

tissues were unsuccessful. Therefore, immunohistochemistry was used to describe the fibrotic

response. AA and CCl4 prOVed to be very different models of chronic liver disease, as one

resulted in massive necrosis (AA) while the other resulted in cirrhosis (CCl4) aS OVidenced of









diluted in an antibody diluent (Invitrogen, Carlsbad, CA), covered (polyethylene class 4, cut just

smaller than the size of the slide) and kept overnight at 40C. The slides were uncovered and

washed with TBS-T 5min then the secondary antibody (1:200) is added 30min. Then washed in

TBS-T 5min, ABC 30 min, wash in TBS-T 5 min and then DAB (in 2.5mL tap water 1 drop

buffer, 1 drop H202, 2 drops DAB) to optimize; H20 1-5 min; Hematoxylin to optimize; H20

15Ssec x 1; Imin x 1; Clarifying solution if necessary up to one min. Bluing optimize; H20 1min;

70% EtOH Imin; 95% EtOH Imin; 100% EtOH Imin x 2; Xylene 1 min x 3; CytosealTM XYL

and coverslipped.

All chromagen stains were run with matching Immunoglobulin controls. Avidin/biotin

solutions (SP-2001), normal serum (Goat, Horse and Mouse), secondary antibodies (goat, rabbit

and mouse), ABC kit (PK-7100), DAB kit (SK-4100) were utilized at room temperature and are

from Vector Laboratories, Burlingame, CA. TBS-T is lX TBS with 0.1% Tween-20.

All samples were photographed using an Olympus microscope and Optronics digital

camera (Olympus, Melville, NY)

Antigen Retrieval

Trypsin digest: 10min 370 (Lab Vision: AP-9008-3A and -lB, Fremont, CA) rinse with

water and resume. Proteinase K digest: 10min 370 (Dako: S3020) rinse with water and resume.

Citra retrieval: pH 6.0, lx Sodium Citrate in H20; 7min in a microwave at 50% power, 18min

cooling (Sigma: S1804), rinse with water and resume.

Immunohistochemistry Analysis

BrdU and a-SMA were analyzed for the number of cells that stained positive. Each were

viewed at 200x magnification and counted in 10 random fields. Days 15, 45 and 90 of AA and

CCl4 treatment were utilized. Hepatocytes and non-hepatocytes were counted for BrdU.

Activated hepatic stellate cells were counted for a-SMA. Microsoft Excel was utilized to find the









Gomori's Trichrome

Tissue sections of 5CLM in size were cut and placed in a 42oC water bath. There were then

lifted from the bath with a SuperfrostTM Plus positively charged slide. The slides were air dried

O/N at RT. The sections were then put into Xylene 5min x 2; 100% EtOH 2min x 2; 95% EtOH

3min; H20 1min; Bouin's (preheated) 1hr 560, rinse in running tap H20 3-8min until yellow

color is removed; Hematoxylin 10min; Rinse in running tap H20 5min; Trichrome 15min; 1%

Acetic Acid 1min; Rinse in distilled H20 30sec; 100% EtOH Imin x 2; Xylene 1min x 3;

Cytoseal/Coverslip (Trichrome kit: Richard Allan: 87021).

Chromagen Staining

Tissue sections of 5CLM in size were cut and placed in a 42oC water bath. There were then

lifted from the bath with a SuperfrostTM Plus positively charged slide. The slides were air dried

O/N at RT. The paraffin was removed and the slides rehydrated by incubating them in Xylene 2

x 5min, 100% Ethanol 2 x 2min, 95% EtOH 2 x Imin, and distilled H20 for 1min. If an antigen

retrieval step is necessary, it is performed at this point then rinsed with distilled water before

proceeding. The slides were then transferred to a slide cassette holder and rinsed with TBS-T

5min. Liver contains endogenous biotin, biotin receptors, and avidin binding sites and in order to

utilize The Vectastain Elite kit (a very sensitive procedure), they must be blocked to prevent

nonspecific staining. Another source of nonspecific staining can occur with the secondary

antibody when this kit can bind to the antigens of the antibody source. Serum blocking was

utilized by taking the serum of the animal that the secondary antibody was made. To conserve

time and the avidin and biotin solutions, they were diluted by performing two serum blocking

steps. Avidin (15ul/ml of 1mL TBS-T with 4 drops avidin) 15min and washed in TBS-T 5min,

then biotin (15uL/mL of 1mL TBS-T with 4 drops biotin) 15min and washed in TB S-T 5min.

The primary antibody is added to the optimum concentration (as listed in the following table)









Nucleic Acid Isolation, Purification and Quantification

RNA Isolation and Analysis

RNA isolation methods included optimizing crude extraction through a homogenizer,

syringe, grinder and/or mallet; varying pH (2-13.4); heat extractions (melting sections on a hot

plate to varying times and/or varying boiling times), changing the percent of chemicals like SDS

from 0. 1% up to 2% and EDTA. RNAzol Bee is used by adding 20C1L of the RNAzol Bee

solution, 80C1L of RNAse free water, 350C1L RLT/Beta-Mercaptinoethanol and 250C1L 100%

EtOH to 30-50C1L of RNA solution. Then inverting the solution before pipetting 700C1L into the

Qiagen Minispin cleanup pink columns. The columns were then centrifuged for 15sec at

10,000RPM and the collection tube was emptied, 500C1L of Buffer RPE was added then

centrifuged for 15sec at 10,000RPM and repeated for 2min. Excess buffer was removed by full

speed centrifugation for 1min. A new tube was used to elute with of RNase-free water in the

amount according to the pellet size (10-50C1L). Then directly frozen at -80oC or put in 55oC water

bath for 10min and inverted then either used or froze at -80oC, or inverted at room temperature

and the concentrations were determined by spectrophotometry. Another method used was a kit

from Ambion, RecoverallThl Total Nucleic Acid Isolation (Ambion, Austin, TX). This kit also

did not yield any product for RNA or DNA.

Reverse Transcription PCR

RNA was converted into cDNA through reverse transcription PCR (rtPCR) in the

following procedure. 1-5Clg of RNA was added to a mixture of 10mm dNTPs (1C1L per 20C1l

reaction), random hexamers (2.5C1L per 20C1L reaction) and water then were incubated at 65oC

for 5min then ice for 1min. Then 5C1L of 10x RT buffer, 0.1Ix DDT and RNase Out are added in a

2:2: 1 respectfully. The samples were mixed and incubated at 42oC for 2min and 1C1L of









which is involved in mineral metabolism and bone growth. The liver utilizes vitamin K for the

production of clotting factors and synthesizing vitamin K-dependent coagulation factors.

Processing large volumes of blood induces the liver to function as a detoxifying organ.

Liver enzymes, such as alcohol dehydrogenase (ADH), cytochrome-P450 (CYP) and isoforms of

uridine diphosphoglucuronate glucuronosyltransferase (UGT) allow for the alteration of

chemical composition of many xenobiotics and their subsequent removal. The conversion of

nonhydrophilic drugs to a more water soluble form aids in their excretion by the kidneys occurs

in the liver. Homeostasis of blood iron levels depends directly on the ability of the liver to store

and metabolize iron.

To perform all of these different functions, the liver requires many different cell types

which include hepatocytes, bile ductular epithelial cells, stellate cells, sinusoidal and vascular

endothelial cells, liver specific macrophages and immune cells [1].

Cells of the Liver

Hepatocytes

Hepatocytes have a life span of 300-400 days and are capable of 86 doublings [6; 7]. Their

division results in compensatory hyperplasia, which restores the liver mass after injury. The

hepatocytes are the functional endocrine and exocrine cells of the hepatic lobule and make up to

80% of cellular mass of the liver. One of the most important functions of hepatocytes is the

production of bile. Gap- and tight-junctions on the sides of hepatocyte membranes help to

distinguish the basolateral from the apical membrane domain. The basolateral domain has

abundant microvilli which proj ect into the space of Disse to maximize surface area to allow

optimum exchange between hepatocytes and blood [1]. Hepatocytes communicate through these

connections with the sinusoids, allowing cross-talk with each other and additional cell types.

Cyclic nucleotides, inositol 1,4,5-trisphosphate, and calcium can enter through gap junctions to









to ductular cells in their distinct isoenzyme profiles intermediate filaments (B SD7, OC2, OC3,

OV-1, and OV-6), extracellular matrix proteins (CK8, 18, 19), enzymes and secreted proteins

(alpha-fetoprotein, y-glutamyl transferase) [36; 37; Nagy 1998]. They also possess hepatocyte

features such as producing albumin. They express some hematopoietic stem cell markers such as

CD-34, c-kit and Thyl as well as Flt-3 [39; 40; 41; 33]. When they are present in the liver they

differentiate toward both hepatic and bile ductular epithelial lineages. When going towards

hepatocyte formation, it was shown that oval cells first differentiate into basophilic small

hepatocytes and then into mature adult hepatocytes [42; 43].

Oval cells may constitute more than 50% of the liver during regeneration caused by

administration of 2AAF/PH; it is thought that the cells form a transit amplifying compartment

that includes undifferentiated progenitors, medially-differentiated transit cells, and newly

differentiated hepatocytes [44]. Oval cells are similar to hepatocytes in that both require growth

factors for cell cycle progression, and both cell types also require a 'priming' process in order to

respond to these stimuli [45]. Oval cells express c-met, the receptor for HGF and therefore are

growth responsive during the time of regeneration when HGF levels are high [46]. Instead of

oval cells simply creating numerous bile ducts, there is morphologic evidence that these cells are

potentially going through a preneoplastic process.

Petersen et. al exposed rats to methylene dianaline (DAPM) 24hrs prior to hepatic damage

(2-AAF/hepatic injury, PH or CCl4) [35]. Under these circumstances the bile ductular epithelium

was destroyed and the oval cell response was severely inhibited. This study was the first to

elucidate a direct association between the requirement of an intact bile ductular epithelium and

the ability to mount an oval cell response. This data, however, does not prove that oval cells arise

from bile ductular cells because the DAPM could have elicited either a direct or indirect toxic





Figure 3-1 Protein and Nucleic Acid analysis of AA and CCl4 archival paraffin embedded
formalin fixed tissue. (A) Western blot analysis of beta actin in fresh normal rat liver
tissue (FNRL), freshly paraffin embedded, formalin fixed normal rat liver (FNRLF), and
archival paraffin embedded, formalin fixed normal rat liver (ANRLF). (B) Coomassie
blue staining was done on 3 different isolation procedures of archival paraffin embedded,
formalin fixed normal rat liver (NAl-3). NAl represents protein with only basic
purification. NA2 has two additional rounds of purification. NA3 has four additional
rounds of purification. (C) PCR amplification of GAPDH in DNA samples from fresh
NRL (FT1-2), archival formalin fixed, paraffin embedded tissue (APl-2) and fresh
formalin fixed, paraffin embedded tissue (FPl1-2). (D) PCR amplification of GAPDH in
purified DNA samples from fresh formalin fixed, paraffin embedded NRL (lane 2),
archival AA treated rat liver (lane 3-9), archival formalin fixed, paraffin embedded NRL
(lane 10), archival CCl4 treated rat liver (lane 11-14).
















































Figure 3 -13 Fibronectin in chronic liver inj ury models by immunohi stochemi stry. (Al1-3) AA
injury at dl5, 45 and 90, respectively. (Bl-3) CCl4 injury at dl5, 45 and 90,
respectively. (C 1-2) NRL. (D) immunoglobulin control. (E-F) Higher magnification
of d90 AA and D90 CCl4, respectively. A-C1 original magnification=200X, C2-F
original magnification=400X.









at later time points in this treatment, activated hepatic stellate cells and ECM proteins are seen

infiltrating the atypical hepatocytes bundle.

An overall evaluation of the oval cell response in these two injury models shows a greater

number of AFP' cells in chronic AA (Figure 3-5). This is consistent with published research

demonstrating that AA is able to recruit liver progenitor cells to the portal zone to affect repair.

Interpretation of this data is complicated by several facts. The first of these is that only a portion

of the oval cell population is AFP We can not rule out the possibility that the CCl4 mOdel

induces a population of oval cells that do not express AFP. Additional markers for oval cells are

known, but antibodies for these markers do not work on formalin Eixed tissue. The second

complicating factor involves the morphology of the AFP' cells seen in later time points of AA

injury. These cells do not display the classical oval nuclei. Histologically, these cells resemble

mature cholangiocytes. Because these cells are generally arranged into ducts that are within a

fibrous matrix, and do not make contact with hepatocytes, we feel that these cells most likely

represent early progeny of oval cells. They likely have moved toward the cholangiocyte lineage,

but retain expression of AFP. One note in this stain is that oval cells are a heterogeneous

population in which not all oval cells express AFP, perhaps these are further differentiated.

Despite the problems with oval cell quantification associated with using AFP as a marker, it may

be safely concluded that each of these models induces a degree of oval cell proliferation.

The molecular microenvironment in regions of injury plays an important role in

mediating the repair of damaged tissue. While not a direct part of this study, inflammatory cells

are known to be the ultimate source of a maj ority of these molecular signals. TGFP is a known

regulator of hepatic stellate cell phenotype. As such, we analyzed slides from each injury model

for TGFP expression by immunohi stochemi stry. The expression pattern for thi s cytokine differed

































Non Hepatocyte Prolife~ration
-AllyiPlcohol
Carbon Tetrachlonlde
120 -
S100 -
" 80 -


nE 40 -
S20 -


dl5 d30 d45 d60 d70 d90
Time Course


60

r 50-
S40-
30-

~20-



dl5 d30 d45 d60
Tirne Course


d70 d90


Figure 3-3 Proliferating (BrdU ) cells over the time course of chronic AA (blue) and CCl4 (pink)
liver injuries.

















































Figure 3-2 Continued















71









stromelysin MMP-3, but enhance the expression of the inhibitors TIMP-1 and TIMP-3 in human

lung fibroblasts, myometrial smooth muscle cells and articular chondrocytes [134; 135]. In

addition, the production of several matrix components, such as collagen and fibronectin, is

induced by TGFPl [125].

Latent TGFP is rapidly taken up by the liver cells including hepatocytes [131; 136]. High

levels of TGFP within the hepatocytes can lead to apoptosis [136]. When exogenous TGFP is

administered following 70% partial hepatectomy, hepatocytes and stellate cells were inhibited

from proliferating, but not the endothelial cells [137]. This is consistent with another report

demonstrating that TGFP can act synergistically with endothelial growth factor (EGF) to either

stimulate or inhibit endothelial cell colony formation in vitro, depending on concentration. TGFP

treated livers eventually recover from partial hepatectomy, as hepatocytes overcome inhibition

and proliferate. Recently, TGFP has been shown to stimulate CTGF expression in hepatocytes

[13 8; Gressner OA, 2007]. TGFP can be inhibited by addition of TGFPRII [139].

Overview of Connective Tissue Growth Factor

During tissue repair and early development, TGFP gene expression is coordinately

regulated with connective tissue growth factor (CTGF) [140]. This is putatively due to the

TGFP-responsive element located within the CTGF promoter [141]. With respect to signaling

molecules, SMADs, ras/MEK/ERK, PKC are required during CTGF induction by TGFP in

cultured mesangial cells [142]. CTGF is a cysteine-rich secretary protein of 36-38 kDa, that

belo gs to the CCN protein family named after CTGF, ysteine-rich 61, and n ehroblastoma

overexpressed proteins [143]. All members of this family exhibit a high degree of amino acid

sequence homology (50-90%) possess a secretary signal peptide at their N-terminus and contain

four distinct protein modules. Module 1 is homologous to the N-terminal cysteine-rich regions of

the six "classic" insulin like growth factor binding proteins (IGFBP-1 to -6) [143], and contains a











75. Shakado S, Sakisaka S, Noguchi K, Yoshitake M, Harada M, Mimura Y, Sata M,
Tanikawa K. Effects of extracellular matrices on tube formation of cultured rat hepatic
sinusoidal endothelial cells. Hepatology 1995;22(3):969-73.

76. Kato S, Otsu K, Ohtake K, Kimura Y, Yashiro T, Suzuki T, Akamatsu N. Concurrent
changes in sinusoidal expression of laminin and affinity of hepatocytes to laminin during
rat liver regeneration. Exp Cell Res. 1992; 198(1):59-68.

77. Carlsson R, Engvall E, Freeman A, Ruoslahti E. Laminin and fibronectin in cell
adhesion: enhanced adhesion of cells from regenerating liver to laminin. Proc Natl Acad
Sci USA. 1981;78(4):2403-6.

78. Gressner AM, Weiskirchen R. Modern pathogenetic concepts of liver fibrosis suggest
stellate cells and TGF-P as maj or players and therapeutic targets. J. Cell. Mol. Med.
2006;10:1:76-99.

79. Bataller R, Brenner DA. Liver fibrosis. The Journal of Clinical Investigation
2005;115:2:209-218.

80. Anthony PP, Ishak KG, Nayak NC, Poulsen HE, Scheuer PJ, Sobin LH. The morphology
of cirrhosis. Recommendations on definition, nomenclature, and classification by a
working group sponsored by the World Health Organization. Clin. Pathol.1978;31:395-
414.

81. Friedman SL, Roll FJ, Boyles J, Bissell DM. Hepatic lipocytes: the principal collagen-
producing cells of normal rat liver. Proc Natl Acad Sci USA. 1985;82(24):8681-5.

82. Verkhratrsky A. Calcium and cell death. Carafoli E and Brini M (eds.), pp. 465-480,
Springer, Netherlands. Calcium Signalling and Disease; (2007).

83. Ferri K, Kroemer G. Organelle-specific initiation of cell death pathways. Nature Cell
Biology 2001;3.

84. Nakagawa. Caspase-12 mediates endoplasmicreticulum-specific apoptosis
andcytotoxicity by amyloid-beta. Let to Nature. 2000;403.

85. Friedman S, Arthur M. Activation of cultured rat hepatic lipocytes by Kupffer cell
conditioned medium: direct enhancement of matrix synthesis and stimulation of cell
proliferation via induction of platelet-derived growth factor receptors. J. Clin. Invest.
1989;84:1780-1785.

86. Li D, Friedman S. Liver Fibrogenesis and the role of hepatic stellate cells: new insights
and prospects for therapy. Gastroenterology and Hepatology 1999; 14(7):618-33.









subj ected to the toxic metabolite and die. Additionally, mature hepatocytes that enter the cell

cycle have been shown to transiently adopt an immature phenotype with respect to xenobiotic

metabolism [167]. This may render hepatocytes in the mid and central zones susceptible to AA

induced damage. The ploidy of the cell determines how quickly the cell can replicate; the fewer

the copies of chromosomes, the more quickly the cell can divide. Since the periportal cells are

damaged, they are not able to repopulate the zone and the mid and pericentral zone cells are left

to repopulate the liver. The closest cells are in the mid zone and have an average ploidy of 16N

which means the cells would have to copy 8 times as many chromosomes as the original

periportal cells. It is however possible to divide without replication and yield two 8N cells, but

this is a less frequent event. The fact that these cells take longer to divide means that a lag time

may occur, and by the time the cells replicate, another dose of AA is given which damages more

cells and the liver may not able to heal.


There may be another reason why we did not see an attempt to heal the liver. A study by

Jung in 2000 reported that a certain dose was needed to yield the proper response (36mg/kg body

weight, given twice weekly for up to 16 weeks) [94]. Jung used Sprauge-Dawely rats which are

known to show different responses to chemicals than the Fischer-344 rats. Our study utilized

7.4mg/kg body weight given 5 days consecutively with 2 days consecutively off for up to 13

weeks. Perhaps their procedures lead to the development of reactive oxygen species which

stimulate the hepatic stellate cells to lay down a protective scar to prevent further exposure of the

chemical to the enclosed area. The most striking aspect of the AA model is the lack of

proliferating hepatocytes (Figure 3-3,4). This phenomenon is difficult to explain, and was

evident in each AA treated animal examined throughout this study. Even though it would take

longer to divide, these hepatocytes should still be proliferating. As mentioned above, it is









paranchymal cells (NPC) and pericentral hepatocytes (Figure 1-4). Unlike hepatocytes that

display a wave of DNA synthesis from periportal to pericentral, NPCs across the lobule exhibit

simultaneous DNA synthesis. The original liver mass is usually restored within 10 days of the

hepatectomy .

Liver regeneration involves a tightly regulation of cells to enter cell cycles after injury and

to exit cell cycles after appropriate tissue remodeling. In the hepatocyte-driven injury response,

almost all types of cells in liver proliferate at least once with some cells dividing two times after

the completion of regeneration. Hepatocytes are the first to enter DNA synthesis in 12hrs after

2/3 PH injury detected with bromodeoxyuridine labeling, followed by HSCs, KCs and then ECs

as seen in figure 1-4. With organ maturity, the process of differentiation leads to the commitment

of differentiated cells to constitutive functions that maintain homeostasis and specialize functions

that serve organisms needs. In the mature livers of all species, proliferation of all cell types

subsides to a low level, thus, the mature liver consists of two types of cells: intermediate cells,

the hepatocytes, which replicate infrequently, but can respond to signals for replication, and

replicating cells, the stem cells, SECs, KCs, and HSCs, bile duct epithelium, and granular

lymphocytes (pit cells).

The earliest event, occurring within one minute after, is a large increase in the blood level

of hepatocyte growth factor (HGF) released from the remodeling ECM [50]. Active HGF then

binds to its receptor c-met to lead the hepatocytes into the cell cycle. The amount of HGF

putatively is in an inverse relationship with total liver mass of hepatocytes. The earliest event,

occurring within one minute after injury, is a large increase in the blood level of HGF released

from the remodeling ECM [50]. Several negative regulatory signals may include TGFP [51],









BIOSKETCH

Alicia Renae Brown was born in 1983 to Gary Ray and Julie Ann (Darr) Brown of Alton,

IL. She was preceded in birth by her sisters Christina Rae, 4 years and less than a day older, and

Andrea Ann by 5 minutes, her fraternal twin sister. They grew up just outside of Brighton, IL, a

small town about an hour north of St. Louis, MO, and attended Brighton Elementary then

Southwestern High School. Alicia advanced to the University of Illinois at Urbana-Champaign in

Animal Sciences with the intent of going into research. While there, she worked in numerous

labs in areas including Reproductive Biology and Endocrinology, Swine Immunology, Maize

Genetics and Nueroimmunology. She has worked directly with respectable professors such as

Drs. Robert Lambert, Humphrey Yao, Janice Bahr and Edward Roy. Her senior year at the

University of Illinois she applied to several schools to obtain a Master's degree in Biology and

she chose the University of Florida based on its top-ranking medical school. At the University of

Florida, she decided to work in the laboratory of Bryon Petersen as the laboratory manager while

she continued her school work.









hepatocytes [29; 42]. Besides hepatocyte lineage, oval cells are also able to differentiate into

intestinal type epithelium in rats in vivo [42], be induced to bile ductular epithelial cells [58] and

pancreatic-like cells in culture [34]. The origins of oval cells remain controversial and several

key questions regarding the molecular cues that initiate oval cell proliferation and direct lineage

specific differentiation still remain.

In progenitor-dependent injury response, although it is not clear whether the same positive

and negative regulatory signals contribute to liver regeneration, injury-induced changes in

cytokines and growth factors definitively modulate the fate of the facultative stem cells (oval

cells) [59]. A distinct response associated with activation of oval cells is a large expansion of

desmin-positive stellate cells and an increased production of growth factors and ECM molecules

in the periportal regions of injured livers. It is believed that stellate cells secrete these growth

factors such as HGF, TGFP, TNFu and ECM components such as fibronectin and regulates

expansion and differentiation of the oval cell population. In addition, the autocrine production of

TGFP, acidic fibroblast growth factor and insulin-like growth factor (IGF) II in oval cells is

suggested to mediate the progenitor-dependent liver regeneration.

Recently, analysis of morphological changes using immunoelectromicroscopy and

immunostaining reveals that activated oval cells form elongated ductular structures, which are

surrounded by basement membranes and terminated at hepatocytes located at the limiting plate

and accompanied by activated and proliferating stellate cells. The surrounded base membranes

may provide a substrate and scaffold for proliferation and migration of oval cells. More

interestingly, a number of proliferating HSCs are always associated with these ductules and

sometimes form direct cell-cell contact with the ductular epithelial cells. This connection









hexachloroethane, and carbon monoxide [1]. This radical reacts with GSH to form GSH-

containing radicals [106] and has been considered to directly act on hepatocytes since 1973

[107]. This formation causes lipid peroxidation in the ER, damage to the plasma membrane and

an increase to intracellular calcium content [108]. Lipid free radicals and peroxides fare less

reactive, and diffuse out to cells throughout the lobule. These radicals can, however, cause

oxidative damage to cell membranes and ER. Lipid peroxidation also results in the production of

aldehydes which require metabolism by ALDH to prevent adduct formation with cellular

macromolecules [108; 109; 110].

Smuckler in 1976 confirmed that CCl4 induces steatosis by disruption of lipid transport.

Synthesis of fat was not increased. He also reported that decreased lipid transport was due to

problems with protein synthesis based on inadequate ribosomal activity. Many hepatocyte

proteins, including albumin, are essential for transporting lipids out of a cell are now reduced

leading to more downstream effects [111].

In 1936 Cameron reported that in acute studies, CCl4 has been shown repeatedly to cause

massive necrosis of the central vein within 24hr of exposure followed by rapid proliferation of

the remaining cells within 5-6d to reach full restoration if it was a one time exposure [1 11].

Chronic exposure leads to a much more serious problem. Repeated exposures lead to more

central vein damage and necrosis, and subsequential periportal damage, micro nodules, excessive

lipid build up (steatosis), abundant connective tissue development, disruption in calcium

homeostasis, streaming cords of fibrous tissue, cirrhosis, oval cell and biliary cell infiltration and

has significant risk of cancer [1; 19; 1 11; 1 12]. In rats treated twice weekly with CCl4 foT fouT

weeks, significant liver fibrosis with septae formation occurred. When CCl4 treatment was

terminated at four weeks, the liver fibrosis resolves completely and normal liver histology was









CHAPTER 4
DISCUSSION

The overall goal of this proj ect was to investigate differences in the fibrotic response to

chronic hepatic portal zone injury versus chronic central zone injury. Deposition of ECM in each

of these injuries is orchestrated by an extensive series of molecular signals, which interact with

inflammatory cells as well as fibroblasts. Classical wound repair theory dictates that the

inflammatory cell component of the response to injury governs the activity of tissue fibroblasts

which, in turn, synthesize ECM as a scaffold on which new cells may migrate and proliferate. In

the context of chronic injury, this process often goes awry resulting in the inappropriate

accumulation of acellular matrix or a fibrotic "scar". Our models of chronic liver injury

demonstrate that this process may be significantly different depending on the location of injury

across the hepatic lobule.

Analysis of Archival Tissue

Initially, we attempted to purify protein suitable for Western blot analysis from archival

formalin fixed, paraffin embedded tissue. This would have benefited the current study by

allowing for quantification of molecular signals and ECM components from the vast archive of

injury models accumulated in this laboratory. The initial challenge was to extract wax free tissue

from paraffin blocks. This proved to be easily accomplished by heating sections of tissue (up to

40Clm) in xylene for up to 1 hour. The fixed tissue proved to be fairly resilient to SDS treatment

and a solution of 2% SDS was required to completely solubilize tissue homogenates. We found

that supernatants from tissues processed in this manner contained protein that could be

precipitated with cold acetone.

We soon discovered that the proteins isolated from these samples were not usable for

Western blot analysis (Figure 3-1 A). It is likely that residual crosslinking of protein, a result of









There is evidence that the TGFPR tetramer is rapidly endocytosed via a clathrin dependant

process. However, this internalization is not required for Smad phosphorylation and is likely a

mechanism for receptor recycling [127; 132].

Biological Activity of Transforming Growth Factor Beta

TGFP has been shown to elicit multiple biological effects dependant on cell type and the

presence of other growth factors. It can either stimulate or inhibit cell proliferation and may

regulate the action of other growth factors [133]. TGFP stimulates the synthesis of maj or

extracellular matrix proteins, including collagen, proteoglycans, glycosaminoglycans, fibronectin

and integrins. TGFP also facilitates ECM deposition by inhibiting the synthesis of matrix

metalloproteinases (MMPs) and inducing the production of tissue protease inhibitors (TIMPs).

This influence on protease activity is species, tissue and cell type specific. TGFP has also been

demonstrated to play a key role in carcinogenesis, as it is able to regulate expression of both

proto-onco genes and tumor suppressor genes.

TGFP has also been shown to function through both the mitogen-activated protein kinase

(MAPK) and the phosphatidylinositol 3-kinase (PI3K) signaling cascades. Together, these

regulate proliferative responses, activating cell cycle progression as well as collagen gene

expression. Smad and MAPK signaling have been found to independently and additively

regulate collagen I gene expression while MAPK, but not Smad signaling increases collagen I

mRNA stability [87]. With respect to liver fibrosis, TGFP mediated activation of Smad 2/3,

appears to be the key signaling pathway [124].

Overexpression of TGFP in rodent models yields an increased expression of protease

inhibitors, such as TIMP and PAl-1 resulting in excessive matrix accumulation. This increase in

protease inhibition coupled with an increased rate of matrix deposition forms the cornerstone of

fibrosis [120]. TGFP has been reported to reduce the production of collagenases MMP-1 and







i



''-:
:;I

~;: '~


.r



"1
''
:


.. Ir
r ~i .r --
.~; Si
I' ~'
--. 'cn~YI~


.~ b~m~G.C'"i' -.-;
I :
r
"
.. ~ '~,
+
~'r ~"'
~.'ur; ~'
I* c,'t

$~r
::

i~~
'

~j~S~Si".2"....) *. ....'1..~



j,.
... (i
1.
r
-t' '' .e
I C
c,. ~J
"";:i-'-r
~' ~j'i
~

:~~ ...
"'' '~. :
: -~"r?.. -'r
:~:5'; '::' ~' :''''''' "

''


V- ~::



-

.'
P;'


*, -.






.
.


Figure 3-4 Immunohistochemical labeling of proliferating, BrdU' nuclei. (Al-3) AA injury at

dl5, 45 and 90, respectively. (Bl1-3) CCl4 injury at dl5, 45 and 90, respectively. (C)

NRL (D) Rat liver, 24hr following partial hepatectomy. Original

magnifications=200X.










rare; oval cells are found scattered throughout the fibrotic cord structures (Figure 3-5F). Also

seen in the CCl4 mOdel are several AFP' transitional hepatocytes are apparent at later time points

(Figure 3-5B).

Trasnsforming Growth Factor Beta Expression by Hepatic Cells in Chronic AA and CCl4
Models

The profibrotic cytokine TGFP was used to identify tissue microenvironments conducive

to the accumulation of ECM. Normal liver demonstrated very mild TGFP expression that was

more evident in the portal zone (Figure 3-6C). TGFP was found to be expressed by a variety of

cell types in each of the chronic injury models. In the CCl4 mOdel, TGFP is strongly expressed

by non-parenchymal cells that are in intimate contact with fibrotic cords. A more diffuse TGFP

staining is evident in the population of small hepatocytes peripheral to the necrotic zone in early

time points (Figure 3-6B1). Expression of TGFP by these cells subsides by middle to late time

points (Figure 3-6B2-3 and F). In the AA model, early time points indicate mild TGFP

expression by hepatocytes within the portal zone (Figure 3-6Al). In later time points, expression

shifts to the non-parenchymal cell fraction within areas of fibrosis (Figure 3-6A2-3). Large areas

of non-specific staining are apparent within the necrotic zones in later AA time points (Figure 3-

6A3).

Connective Tissue Growth Factor Expression by Hepatic Cells in Chronic AA and CCl4
Models

CTGF has been shown to be a downstream mediator of TGFP expression and was used as

an additional indication of a pro-fibrotic microenvironment (70). Early time points in the AA

model shows punctate expression of CTGF by individual hepatocytes within the central zone

(Figure 3 -7Al1). Expression of this factor increases to involve a maj ority of the central zone

hepatocytes in the mid time points and eventually decreases by later time points (Figure 3-7A2-

3). Also, occasional non-parenchymal cells within areas of fibrosis are positive for CTGF. This










paraffin. All paraffin sections were cut to a 5Cpm thickness. Liver sections were collected at the

following time points: day 15, 30, 40, 45, 55, 60, 65, 70 and 90.

Protein Isolation, Purification and Quantification

Fresh Normal Rat Liver Extraction

Protein was isolated from formalin fixed paraffin embedded (FFPE) rat liver tissue.

Controls used were snap frozen normal rat liver (NRL) tissue, fresh FFPE NRL and archival

FFPE NRL. Whole liver sections from snap frozen NRL were cut into small pieces (about 150-

200Clg) and added to 10ml of RIPA buffer (50mM Tris-Cl pH 7.4, 250mM NaC1, 5mM EDTA,

50mM NaF) with proteinase inhibitors (per 1mL of RIPA buffer 30C1l Aprotinin, 10CIL ImM

leupeptin, 10CIL 100mM phenylmethylsulfonyl fluoride (PMSF) and 10CIL Na orthovandate

were added). The fresh tissue is kept on ice for the remainder of the procedures until it is ready

for analysis. The fresh tissue is disrupted utilizing either a homogenizer (10sec at a time until the

proper consistency is reached) or a needle (18-26.5 gauge) and syringe where it is pipetted up

and down until tissue was thoroughly homogenized. The sample was vortexed for 30sec and then

centrifuged at 10,000xg at 4oC for 10min to remove excess lipids and DNA. The supernatant was

collected into 2.0mL screw cap tube and place in -80oC until use.

Formalin Fixed Paraffin Embedded Tissue Extraction

For FFPE tissues, preceding methods were necessary due to paraffin and formalin

contaminations. The following was performed individually and in different combinations.

Tissues were either cut on a microtome or a portion of the liver section was dissected out with a

scalpel. They were then subj ected to paraffin removal via heat and or xylene treatments. The

buffer solution (RIPA and proteinase inhibitors) was altered through chemical composition,

protease inhibitors and pH. More stringent disruption methods were added. Cleanup was also











151. Wunderlich K, Senn BC, Todesco L, Flammer J, Meyer P. Regulation of connective
tissue growth factor gene expression in retinal vascular endothelial cells by angiogenic
growth factors. Graefes Arch Clin Exp Ophthalmol. 2000;238(11):910-5.

152. Pendurthi UR, Allen KE, Ezban M, Rao LV. Factor VIIa and thrombin induce the
expression of Cyr6 1 and connective tissue growth factor, extracellular matrix signaling
proteins that could act as possible downstream mediators in factor VIIa x tissue factor-
induced signal transduction. J Biol Chem. 2000 12;275(19):14632-41 .

153. Chambers RC, Laurent GJ. Coagulation cascade proteases and tissue fibrosis. Biochem
Soc Trans. 2002;30(2):194-200.

154. Hahn A, Heusinger-Ribeiro J, Lanz T, Zenkel S, Goppelt-Struebe M. Induction of
connective tissue growth factor by activation of heptahelical receptors. Modulation by
Rho proteins and the actin cytoskeleton. J Biol Chem. 2000;275(48):3 7429-3 5.

155. Abraham DJ, Shiwen X, Black CM, Sa S, Xu Y, Leask A. Tumor necrosis factor alpha
suppresses the induction of connective tissue growth factor by transforming growth
factor-beta in normal and scleroderma fibroblasts. J Biol Chem. 2000;275(20):15220-5.

156. Bradham DM, Igarashi A, Potter RL, and Grotendorst GR. Connective tissue growth
factor: a cysteine-rich mitogen secreted by human vascular endothelial cells is related to
the SRC-induced immediate early gene product CEF-10. J Cell Biol. 1991;114: 1285-
1294.

157. Nishida T, Nakanishi T, Asano M, Shimo T, Takigawa M. Effects of CTGF/Hes24, a
hypertrophic chondrocyte-specific gene product, on the proliferation and differentiation
of osteoblastic cells in vitro. J Cell Physiol. 2000; 184(2):197-206.

158. Wandji SA, Gadsby JE, Barber JA, Hammond JM. Messenger ribonucleic acids for
MAC25 and connective tissue growth factor (CTGF) are inversely regulated during
folliculogenesis and early luteogenesis. Endocrinology 2000; 141(7):2648-57.

159. Igarashi A, Nashiro K, Kikuchi K, Sato S, Ihn H, Fujimoto M, Grotendorst GR, Takehara
K Connective tissue growth factor gene expression in tissue sections from localized
scleroderma, keloid, and other fibrotic skin disorders. J Invest Dermatol.
1996;106(4):729-33.

160. Frazier KS, Grotendorst GR. Expression of connective tissue growth factor mRNA in the
fibrous stroma of mammary tumors. Int J Biochem Cell Biol. 1997;29(1):153-61.

161. Graness A, Poli V, Goppelt-Struebe M. STAT3-independent inhibition of
lysophosphatidic acid-mediated upregulation of connective tissue growth factor (CTGF)
by cucurbitacin I. Biochem Pharmacol. Jun 2006;72(1):32-41.









CHAPTER 3
RESULTS

Optimizing Protein and Nucleic Acid Extractions from Archival Formalin Fixed Paraffin
Embedded Tissue

Despite repeated attempts at purification, electrophoretic resolution of discrete protein

bands from archival tissue was not possible (Figure 3-1B). However, we attempted to resolve

beta actin on Western blots of archival tissue. Fresh NRL displayed a strong band of the

appropriate molecular weight (Figure 3-1A). There was no corresponding band in the archival

tissue lane (Figure 3-1A). Interestingly, fresh formalin fixed NRL also failed to produce a beta

actin band by Western blot (Figure 3-1A).

Amplification of a region of the GAPDH gene in DNA isolated from archival, paraffin

embedded NRL was possible (Figure 3-1C). However, PCR amplification of DNA isolated from

archival, paraffin embedded allyl alcohol (AA) and carbon tetrachloride (CCl4) treated liver was

less successful (Figure 3-1D). As double stranded DNA is much more stable than single stranded

RNA, RT-PCR was not attempted on archival tissue.

Assessing the Allyl Alcohol and Carbon Tetrachloride Chronic Liver Injury Models

Tissue from formalin fixed, paraffin embedded blocks were cut into 5Clm sections and

stained. Time points from days 15, 30, 40, 45, 54, 60, 65, 68, 70 and 90 were used. Days 15, 30,

45, 60, 70 and 90 are shown in H&E. Immunohistochemical data are presented from dl5, 45 and

90. All chromagen stains were run with normal rat liver (NRL) and a matching immunoglobulin

controls. The samples were photographed using an Olympus microscope and Optronics digital

camera Olympus, Melville, NY).









of these proteins results in a fibrotic scar that diminishes liver function. When this progresses to

cirrhosis, the life of the organism is j eopardized. Each of the aforementioned matrix proteins

were assessed in our chronic injury models. Collagen represents the most abundant protein found

in the ECM. In both the AA and CCl4 mOdels, collagen progressively accumulates at the zone of

injury (Figures 3-10,11). Trichrome staining shows that the collagen that accumulates in the

periportal zone in the AA model is much less densely arranged than in the CCl4 mOdel (Figure 3-

10 Al-3). This is an interesting observation as the AA model shows very little accumulation of

collagen IV, which is known to crosslink collagen I into tighter bundles (Figure 3-11 Al-3).

Laminin is also much less abundant in the AA model than the CCl4 mOdel. Laminin has been

shown to bind to collagen IV, so it is not surprising that very little laminin is present in the

injured zones of the AA model which are also devoid of collagen IV (Figure 3-12 Al-3). In the

mid time points, many regions of fibronectin deposition may be seen radiating from the portal

zone into healthy hepatic parenchyma (Figure 3-13 A2). At these same time points, many AFP+

oval cells are present in the portal zone. Our laboratory has previously shown that a fibronectin

rich provisional matrix is associated with proliferating oval cells, and we suspect this is driving

the deposition of fibronectin in the AA model.

Overall, it appears that collagen I is produced by stellate cells at the zone of injury in

each of the models included in this study. Increased profibrotic cytokine activity in the

pericentral, CCl4 injury model correlates with robust stellate cell activation within the damaged

zone. These activated fibroblasts secrete a dense matrix of collagen I that is cross-linked by

collagen IV which is bound with laminin. However, matrix modification by collagen IV occurs

only in the CCl4 mOdel resulting in a much tighter matrix as compared to the AA model. The

dense nature of the fibrotic scar may prevent remodeling by proteinases allowing this ECM to









necessary and performed through acetone precipitation(s). Samples were analyzed through a

modified Lowry assay from Bio-Rad (DC protein assay, Bio-Rad, Hercules, CA).

Amounts and sizes of tissues were varied by cutting the samples on a microtome from 5-

40Clm thick, yielding 5 to 500Clm total, or by removing a piece with a scalpel from about 150-

200Clg. This piece is subj ected to either going straight into the next step or is broken down into

smaller pieces with a mallet. Tissues were heated to varying times and temperatures to allow

paraffin to melt and assist in breaking the crosslinking. Heating was performed on aluminum foil

upon a hot plate or inside a plastic heat sealed bag in boiling or near boiling (900C) water.

Paraffin removal through xylene was performed for 5min x 2 to 30min x 2 followed by

dehydration through 100% ethanol (EtOH) 5min then 30 min, 90% EtOH for 5min then 30min,

70% EtOH for 5min and then into a buffer solution. In the buffer solution, the tissue was either

transferred into a plastic bag and disrupted with a mallet, or into a tube to be homogenized as

described for normal tissue.

Buffer solutions were altered through a variety of techniques. Altering the pH was the first

attempt, ranging from a very acidic 2.0 to a highly caustic 13.4. The next option in altering the

buffer was changing the percent of chemicals such as SDS from 0% up to 2% (typical is 0. 1%)

and EDTA (from 1mM to 10mM with 5mM being standard). Citrate buffer was added to the

solution to aid in the removal of protein crosslinking.

Then protein handling was considered. The proteins typically are used the same day as

extraction and are not frozen. Freeze/thaw cycling was performed from one to three times.

Boiling was also performed at varying times (0-20min) in the varying solutions for crosslink

removal.









Zone 2 cells which are typically 16N. Conversely, hepatocytes within Zone 1 are smaller and

usually single nucleated (2N) [5].

The many functions of the liver involve maintaining homeostasis of the body, metabolizing

amino acids (forming ammonia and converting it to urea), lipids, xenobiotics, serum proteins and

carbohydrates, converting glucose into glycogen for storage. The liver is also one of the main

sites for insulin and glucagon degradation which assists the pancreas control blood glucose

levels. In animals, regulation of blood glucose levels by the liver is a vital part of homeostasis. In

hepatocytes, any extra glucose is stored as glycogen through phosphorylation into glucose-6-

phosphate (G6P) and, when needed, glucose can be released into the bloodstream through

G6Pase.

In addition, the liver maintains the colloid osmotic pressure of the blood by producing the

most abundant plasma protein, albumin. It also produces other vital plasma proteins such as

lipoproteins, glycoproteins including prothrombin and fibrinogen, and the nonimmune a- and P-

globulins. Although the liver does not actively produce hormones, it modifies the actions of

hormones released by other organs. The liver stores and converts vitamins A, D and K taken up

from the blood stream to a more functional form. Stellate cells store vitamin A as retinyl esters

within their lipid pools. Vitamin A is important for vision as it is a component of rhodopsin (a

pigment in the rods and cones of the eye) and for proper bone growth among others. Vitamin D3

cholecalciferoll) is generated in the skin of animals when light energy is absorbed by a precursor

molecule 7-dehydrocholesterol Cholecalciferal is hydroxylated to 25-hydroxycholecalciferol by

the enzyme 25-hydroxylase (CYP27, which is found in high concentrations in the liver) then it

goes to the kidney to be further modified into the active form of 1,25-dihydroxycholecalciferol









87. Tsukada S, Westwick JK, Ikejima K, Sato N, Rippe RA. SMAD and p38 MAPK
signaling pathways independently regulate alphal (I) collagen gene expression in
unstimulated and transforming growth factor-beta-stimulated hepatic stellate cells. J Biol
Chem. 2005;280(11): 1005 5-64.

88. Gressner OA, Lahme B, Demirci I, Gressner AM, Weiskirchen RJ. Differential effects of
TGF-beta on connective tissue growth factor (CTGF/CCN2) expression in hepatic stellate
cells and hepatocytes. Hepatology 2007;47(5):699-710.

89. Bissell DM. Chronic liver injury, TGF-b, and cancer. Experimental and Molecular
Medicine 2001;33:4: 179-190.

90. Blom IE, Goldschmeding R, Leask A.Bork P. Matrix Gene regulation of connective
tissue growth factor: new targets for antifibrotic therapy? Biol. 2002;21(6):473-82.

91. Badr MZ, Belinsky SA, Kauffman FC, and Thurman RG. Mechanism of hepatotoxicity
to periportal regions of the liver lobule due to allyl alcohol: role of oxygen and lipid
peroxidation. J Pharmacol Exp Ther 1986;238:1138-1142.

92. Boyland E and Chasseaud LF. Enzyme-catalysed conjugations of glutathione with
unsaturated compounds. Biochem. J. 1967;104:95.

93. Serafini-Cessi F. Conversion of allyl alcohol into acrolein by rat liver. Biochem. J.
1972;128:1103-1107.

94. Jung SA, Chung YH, Park NH, Lee SS, Kim JA, Yang SH, Song IH, Lee YS, Suh DJ,
Moon IH. Experimental model of hepatic fibrosis following repeated periportal necrosis
induced by allylalcohol. Scand J Gastroenterol. 2000;35(9):969-75.

95. Ohno Y, Ormstad K, Ross D, Orrenius S. Mechanism of allyl alcohol toxicity and
protective effects of low-molecular-weight thiols studied with isolated rat hepatocytes.
Toxicol Appl Pharmacol. 1985;78(2):169-79.

96. Hales CA, Barkin PW, Jung J, Trautman E, Lamborghini D, Herrig N and Burke J.
Synthetic smoke with acrolein but not HCI produces pulmonary edema. J Appl Physiol.
1988;64:1121.

97. Ayer HE and Yeager DW. Irritants in cigarette smoke plumes. Am J Public Health
1982;72:1283-1299.

98. Kehrer JP, Biswal SS. The molecular effects of acrolein. Toxicol. Sci. 2000;57: 6-15.

99. Vasiliou V, Pappa A, Estey T. Role of human aldehyde dehydrogenases in endobiotic and
xenobiotic metabolism. Drug Metab Rev. 2004;36(2):279-99.












Laminin Expression by Hepatic Cells in Chronic AA and CCl4 Models .............. .... ...........66
Fibronectin Expression by Hepatic Cells in Chronic AA and CCl4 Models ..........................67

4 DI SCUS SSION ............ ...... ..._. ...............8 3...


Analysis of Archival Tissue.. .................. .......... ........... ............. .................83
Immunohistochemical Analysis of Archival Tissue ................. ...............................84

5 FUTURE DIRECTIONS .............. ...............92....


LIST OF REFERENCES ................. ...............93........... ....


BIO SKETCH ................. ...............107......... ......

















































Figure 3-10 Trichrome staining for collagen I during chronic liver injury. (Al-3) AA injury at
dl5, 45 and 90, respectively. (Bl-3) CCl4 injury at dl5, 45 and 90, respectively. (C1-2)
NRL. (D-E) Higher magnification of d90 AA and D90 CCl4, respectively. A and B
original magnification=600X, Cl original magnification=200X, C2-F original
magnification=600X.









Proteins of the Extracellular Matrix


Collagens

Collagen is the maj or protein in mammals, making up to 25% of all proteins in the body. It

exists in many forms to support its unending set of functions. The uniting structural point of all

collagens is that almost every third residue is glycine. The maj or functions of collagen are tensile

strength, flexibility, motility, adhesion and overall tissue support [19]. The slightest of mutations

in any of the collagens result in maj or defects such as osteogenesis imperfecti. The chief cell

type that produces collagen in the liver is the HSC [21]. Other liver cells have been shown to

produce small amounts of it as well, including hepatocytes and ECs [19]. Collagen accumulation

protects the rest of the organ from spreading viruses and disease, and can provide a scaffold for

regenerating liver cells. However, this accumulation also can cause the cells within to die due to

lack of nutrients. Accumulation in the hepatic sinusoids obstructs blood flow which can also lead

to portal hypertension, edema and ascites. Scarring leads to blocked blood flow, decreasing

function/processing of nutrients, hormones, drugs, toxins and viruses [21]. Since 1986, TGFP has

been known to be an intrinsic regulator of collagens [68]. Collagens are dramatically increased

upon chronic injury from chemicals such as CCl4 and AA [19]. The most abundant protein in the

mammalian body is the fibrillar collagen type I (Coll). Coll transcripts contain CCAAT binding

sequences that are highly conserved between species [69]. CAAT binding proteins and other

transcription factors present in the liver bind these sequences. According to Yang et. al, Coll can

serve as a chemoattractant stimulus for HSCs [65].

Collagen Type IV

Collagen type IV (CollV) is non fibrillar collagen composed of at least six different a

chains which form a sheet (mesh-like) structure. The most common form of this collagen

molecule is the ubiquitous heterotrimer. Its molecules interact to form a complex network via









Stacking Gel

The gels were pored using electronic or plastic Pasteur pipettes to slightly below top of

glass and the comb inserted. After 10-15min, the gel was ready to run after gently removing

comb and rinsing the wells with Milli-Q H20.

Western Blot Analysis

Protein Sample Preparation

The amount of protein to be loaded per well was determined based on the source of

isolation, archival or fresh FFPE, and the sensitivity of the antibody being used for detection.

Samples were added to the appropriate amount of buffer and placed in a screw cap 2.0mL tube.

5C1L of 5X Western Loading Buffer per lane was added to each tube. Each sample was boiled for

0-20min and then incubated at RT for 5min to cool. The samples were immediately loaded and

any remaining solution placed on ice and returned to storage at -800C. Each well of 0.75-1.5mm

gel was loaded with 15-40C1L.

Electrophoresis of the Western Gel

The gel was loaded in the running apparatus with small plate facing inward. The inner

chamber was filled with lX Running Buffer till full and over flows and the outer chamber was

filled 2.0 inches. 15-40pIL of samples was loaded per well with 3 CL of Protein Standard

(kaleidoscope, Bio-Rad) within the appropriate well. Any empty lanes were filled with 5Cl1 of 5X

Western Loading Buffer. The gel was run at 60-80 Volts (based on individual set up flow rates of

the bubbles) until the loading die had migrated out of the stacking gel. Then the gel was run at 90

Volts until the loading die ran the length of the gel.

Transferring of a Western Gel to a PVDF Membrane

The upper left corner of the Immuno-Blot@ PVDF (Bio-Rad) membrane was cut and the

membrane was labeled with pencil. It was then dipped in methanol and soaked in water for five










LIST OF FIGURES


Figure page

1-1 Hepatic microarchitecture and blood flow............... ...............47..

1-2 Liver acinus............... ...............47.

1-3 Growth of remaining three liver lobes after %/ partial hepatectomy in the rat.. ........._......47

1-4 Amount of resident hepatic cells within the cell cycle during the time following %/
partial hepatectomy ................. ...............47.......... ......

3-1 Protein and Nucleic Acid analysis of AA and CCl4 archival paraffin embedded
formalin fixed tissue .............. ...............69....

3-2 Hematoxylin and Eosin staining of AA and CCl4 treated rat livers 90 days following
initiation of injury ................ ................ ....___ ....._ ....___ ....._ .70

3-3 Proliferating cells over the time course of chronic AA and CCl4 liVeT injurieS ...............72

3-4 Immunohistochemical labeling of proliferating, BrdU' nuclei ................. ............... ....73

3-5 Alpha Fetal Protein' hepatic oval cells in the portal regions of rat liver ................... ........74

3-6 Transforming Growth Factor Beta expression during chronic liver injury .......................75

3-7 Connective Tissue Growth Factor expression during chronic liver injury ................... .....76

3-8 Alpha-Smooth Muscle Actin expression during chronic liver injury ............... .... ........._..77

3-9 Activated HSCs over the time course of chronic AA and CCl4 liVeT injurieS ...................78

3-10 Trichrome staining for collagen I during chronic liver injury .............. .....................7

3-11 Collagen type IV in chronic liver inj ury models by immunohi stochemi stry ................... ..80

3-12 Laminin deposition in chronic liver injury models by immunohistochemi stry .................81

3-13 Fibronectin in chronic liver injury models by immunohistochemistry ................... ...........82









matrix by breaking down existing ECM as new ECM is produced [19]. Degradation of existing

matrix does not necessarily limit fibrosis, as numerous growth factors adhered to the

extracellular matrix are liberated, creating a new and progressive microenvironment through

which the fibrogenic message is passed [120]. The balance between these two competing HSC

functions determines whether or not fibrosis occurs.

Overview of the Transforming Growth Factor Beta

There are over 100 distinct members of the TGFP superfamily sharing at least one

conserved amino acid domain [121]. This protein was discovered by Roberts in 1983 while

conducting studies on fibroblasts that had been retrovirally transformed in vitro. This growth

factor has been described as both heat and acid stable, and may be extracted from a wide variety

of cells [122]. TGFP is a 25 kD cytokine that inhibits epithelial cell and lymphocyte

proliferation. Conversely, it is able to stimulate proliferation and extracellular matrix production

by fibroblasts [89; 123]. TGFP is also known to play a key role in tissue remodeling and in

immunological self-tolerance [124]. In addition to being directly secreted by cells, a reservoir of

TGFP is covalently bound to the N-terminus of collagen IIA [125].

Significant production of this cytokine occurs within hepatic fibroblasts and macrophages,

and represents a maj or component of the secretary vesicles of platelets. Work by Carr in 1986

showed that TGFP is inhibited during severe liver injury, allowing cell proliferation. In cultures

of primary rat hepatocytes, TGFP bound its receptor with an apparent Kd of 93. 1 pM. This

binding constant was found to be transiently diminished following partial hepatectomy, returning

to normal by 96 hours [126].

Three members of the TGFP family (TGFP l, 2 and 3) have been identified in mammals.

TGFP4 and 5 are found in the chicken and Xenopus respectively. TGFP requires proteolytic

cleavage to an active form before it will bind to its receptor. Cleavage also mediates the release









KC Kupffer cell

LM Laminin

LRP Low density lipoprotein receptor

MMP Matrix metalloproteinase

MAPK Mitogen-activated protein kinase

NASH Nonalcoholic steatohepatitis

NF-icB Nuclear factor kappa B

PH Partial hepatectomy

PDGF Platelet-derived growth factor beta

PI3K Phosphatidylinositol 3-kinase

SEC Sinusoidal endothelial cell

TIMP Tissue inhibitor of metalloproteinases

TGFP Transforming growth factor beta

TGFPR Transforming growth factor beta receptor

TNFu Tumor necrosis factor alpha

vWF von Willebrand factor
















































Figure 3-6 TGFP expression during chronic liver injury. (Al-3) AA injury at dl5, 45 and 90,
respectively. (Bl1-3) CCl4 injury at dl5, 45 and 90, respectively. (C1-2) NRL. (D)
immunoglobulin control. (E-F) Higher magnification of d90 AA and d90 CCl4,
respectively. A and B original magnification=600X, Cl original magnification=200X,
C2-D original magnification=600X, E-F original magnification=200X.










LIST OF TABLES


Table page

2-1 List of antibodies utilized............... ...............60

3-1 Proliferating hepatocytes and non-hepatocytes over the time course of chronic AA
and CCl41Ve T injurieS ................. ...............68................

3-2 Activation of HSCs over the time course of chronic AA and CCl4 ................. ................68









COMPARISON OF EXTRACELLULAR MATRIX PROFILES INT ALLYL ALCOHOL AND
CARBON TETRACHLORIDE CHRONIC LIVER INJURY MODELS




















By

ALICIA RENAE BROWN


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA


2008












Experimental Animals ................ ..... ...............48.
Animal Sacrifice and Tissue Collection .............. ...............48....
Protein Isolation, Purification and Quantification ................. ...............49......._.. ...
Fresh Normal Rat Liver Extraction ................... ......__. ...............49....
Formalin Fixed Paraffin Embedded Tissue Extraction .............. ....................4
Acetone Precipitation ............... ...... ... ..............5
Protein Quantification with DC Protein Assay .............. ...............51....
Running Gel............... ...............51..
Stacking Gel .............. ...............52....
W western Blot Analysis .............. ...............52....
Protein Sample Preparation ................. ...............52...............
Electrophoresis of the Western Gel ............... ... ........... ...............52.....
Transferring of a Western Gel to a PVDF Membrane .............. ...............52....
Coomassie Staining of Western Gel ................. ...............53........... ...
Silver Staining of Western Gel ................. ...............53.......... ....
Probing of Western Membrane ................... ... ... ........ ...............54......
Developing of Western Membrane with ECL PlusTM. . . ....................... .. .. .. .. .. .. .54
Membrane Stripping for Reprobing .............. .... ...............54..
Nucleic Acid Isolation, Purification and Quantification .............. ...............55....
RNA Isolation and Analysis............... ...............55
Reverse Transcription PCR ................. ...............55..._..._......
DNA Isolation and Analysis............... ...............56
PCR Analysis............... ...............56
Histological Analysis............... .. ....... ...............5
Hematoxylin and Eosin Staining of Paraffin Embedded Tissue ................ ........._......57
Gomori's Trichrome............... ...............5
Chromagen Staining ................. ...............58.......... .....
Anti gen Retrieval ................. ...............59.......... ......
Immunohistochemistry Analysis .............. ...............59....

3 RE SULT S .............. ...............61....


Optimizing Protein and Nucleic Acid Extractions from Archival Formalin Fixed
Paraffin Embedded Tissue ................... .. ..... .. ........................6
Assessing the Allyl Alcohol and Carbon Tetrachloride Chronic Liver Injury Models..........61
Optimizing Protein and DNA Extractions from Archival Formalin Fixed Paraffin
Embedded Tissue .............. ................ .........6
Overall Evaluation of the Two Injury Models. .............. ..... ....... .......... ................. .....6
Hepatocyte and Non-Parenchymal Cell Proliferation in Chronic AA and CCl4 Models ......63
Alpha Fetoprotein Expression by Hepatic Oval Cells in Chronic AA and CCl4 Models.......63
Trasnsforming Growth Factor Beta Expression by Hepatic Cells in Chronic AA and
CC l4 M odels.........._ _. ........... _... .._._ ... ..... .. .. ... ... ........6
Connective Tissue Growth Factor Expression by Hepatic Cells in Chronic AA and CCl4
M odels .........._.... ........._._ ... ..._ ... .. ......... .. .............6
Alpha-Smooth Muscle Actin Expression by Hepatic Stellate Cells in Chronic AA and
CCl4 M odels .........._..... ...... .. ._._ ...... .._ ...... ......... .........6
Collagen Expression by Hepatic Cells in Chronic AA and CCl4 Models .............. ................65









63. Paralkar VM, Vukicevic S, Reddi AH. Transforming growth factor beta type 1 binds to
collagen IV of basement membrane matrix: implications for development. Dev Biol.
1991;143(2):303-8.

64. Zarnegar R, Michalopoulos GK. The many faces of hepatocyte growth factor: from
hepatopoiesis to hematopoiesis. J Cell Biol. 1995;129(5):11 77-80.

65. Yang C, Zeisberg M, Mosterman B, Sudhakar A, Yerramalla U, Holthaus K, Xu L, Eng
F, Afdhal N, Kalluri R. Liver fibrosis: insights into migration of hepatic stellate cells in
response to extracellular matrix and growth factors. Gastroenterology 2003;124(1): 147-
59.

66. Skalli O, Pelte MF, Peclet MC, Gabbiani G, Gugliotta P, Bussolati G, Ravazzola M, Orci
L. Alpha-smooth muscle actin, a differentiation marker of smooth muscle cells, is present
in microfilamentous bundles of pericytes. J Histochem Cytochem. 1989;37(3):3 15-21.

67. Hautmann MB, Madsen CS, Owens GK. A transforming growth factor beta (TGFbeta)
control element drives TGFbeta-induced stimulation of smooth muscle alpha-actin gene
expression in concert with two CArG elements. J Biol Chem. 1997;272(16): 10948-56.

68. Roberts AB, Sporn MB, Assoian RK, Smith JM, Roche NS, Wakefield LM, Heine UI,
Liotta LA, Falanga V, Kehrl JH, et al. Transforming growth factor type beta: rapid
induction of fibrosis and angiogenesis in vivo and stimulation of collagen formation in
vitro. Proc Natl Acad Sci U S A. 1986;83(12):4167-71.

69. Bornstein P, Sage H. Regulation of collagen gene expression. Prog Nucleic Acid Res Mol
Biol. 1989;37:67-106.

70. Pi L, Oh SH, Shupe T, Petersen BE. Role of connective tissue growth factor in oval cell
response during liver regeneration after 2-AAF/PHx in rats. Gastroenterology
2005;128(7):2077-88.

71. Gui Y, Murphy LJ. Insulin-like growth factor (IGF)-binding protein-3 (IGFBP-3) binds
to fibronectin (FN): demonstration of IGF-I/IGFBP-3/fn ternary complexes in human
plasma. J Clin Endocrinol Metab. 2001;86(5):2104-10.

72. Wewer UM, Engvall E, Paulsson M, Yamada Y, Albrechsten R. Laminin A, Bl, B2, S
and M subunits in the postnatal rat liver development and after partial hepatectomy. Lab
Invest. 1992;66(3):378-89.

73. Fox JW, Mayer U, Nischt R, Aumailley M, Reinhardt D, Wiedemann H, Mann K, Timpl
R, Krieg T, Engel J, et al. Recombinant nidogen consists of three globular domains and
mediates binding of laminin to collagen type IV. EMBO J. 1991;10(11):313 7-46.

74. McKee KK, Harrison D, Capizzi S, Yurchenco PD. Role of laminin terminal globular
domains in basement membrane assembly. J Biol Chem. 2007;282(29):21437-47.









Superscript II was added to each and continued at 42oC for another 50min until terminated at

70oC for 15min. 1CIL of RNase H was added and incubated at 37oC for 20min to rid the samples

from any residual RNA. All products are from Invitrogen's Superscript II kit. The cDNA was

processed as DNA as listed in PCR Analysis.

DNA Isolation and Analysis

Altered methods utilized included optimizing crude extraction through a homogenizer,

syringe, grinder and/or mallet; varying pH (2-13.4); heat extractions (melting sections on a hot

plate to varying times and/or varying boiling times), changing the percent of chemicals like SDS

from 0. 1% up to 2% and EDTA. The main issue is contamination of cellular debris and formalin

cross-linking, so the tissues were isolated and washed multiple times. Isolation and washing were

also varied. Samples are reisolated through Promega's DNA Wizzard, Sigma's PCI mix, or plain

phenol, then are mixed via vortexing or inverting for a range of times (1min-4hrs) then

centrifuged at 14000xg for 10-15min at 4oC. The top clear supernatant is removed with or

without the white protein layer while discarding the bottom phenol layer. Either isopropanol (at

1:1) or 100% ethanol (1:2.5) is added with 1/10th VOlume of 3M sodium acetate pH 5.2 to

visualize DNA in solution or pelleting. This continued until the protein layer was no longer

visible which decreased the concentration, but more efficiently purified the DNA for PCR

analysis.

PCR Analysis

DNA concentrations were measured with a spectrophotometer but they were not

completely accurate due to the contamination problems faced with utilizing fibrotic, archival

formalin fixed paraffin embedded tissues. A substantial amount of DNA was lost with multiple

isolations and washings. PCR was performed with 0.5-5 Cg of DNA, GAPDH primers at varying









































O 2008 Alicia Renae Brown

















































Figure 3-8 Alpha-Smooth Muscle Actin (a-SMA) expression during chronic liver injury. (Al-3)
AA injury at dl5, 45 and 90, respectively. (Bl-3) CCl4 injury at dl5, 45 and 90,
respectively. (C 1-2) NRL. (D) immunoglobulin control. (E-F) Higher magnification
of d90 AA and D90 CCl4, respectively. A and B original magnification=600X, Cl
original magnification=200X, C2-D original magnification=600X, E-F original
magnification=200X.









is especially true at the later time points (Figure 3-7E). In contrast to the AA model, CTGF

expression in the CCl4 mOdel continually increases over the time course of injury (Figure 3-7

Bl-3). Analysis of high magnification d90 slides demonstrates CTGF expression both by

hepatocytes as well as non parenchymal cells (Figure 3-7F). NRL showed little staining

concentrated mainly within the endothelial cells associated with the hepatic circulatory system

(Figure 3-7C).

Alpha-Smooth Muscle Actin Expression by Hepatic Stellate Cells in Chronic AA and CCl4
Models

We used a-SMA to identify activated HSCs in each chronic liver injury model. As

expected, NRL strongly expresses a-SMA in vessel walls and in occasional HSCs (Figure 3-

8C1-2). Both the AA and CCl4 injury models demonstrate increasing a-SMA expression by cells

associated with areas of fibrosis throughout the time course of these studies (Figure 3 -8Al-3,Bl1-

3). a-SMA expression appears to be confined to within the fibrotic regions in the AA model

while many a-SMA+ cells are evident within the regenerating nodules found in later time points

of the CCl4 mOdel (Figure 3-8 E-F). Quantification of the a-SMA' cells within each of these

models demonstrates a significantly greater number of HSCs in the CCl4 mOdel as compared to

the AA model (Figure 3-9 and Table 3-2).

Collagen Expression by Hepatic Cells in Chronic AA and CCl4 MOdels

Molecular signaling through TGFP and CTGF promote collagen synthesis and deposition

by HSCs. Interstitial collagen types I, II and III (Figure 3-10) and type IV (Figure 3-11) were

identified within the fibrotic areas in both the AA and CCl4 injury models. Trichrome staining

demonstrated an initial increase in periportal, interstitial collagen between the early and middle

time points (Figure 3-10 Al-2). As the injury proceeds, little additional collagen seems to be

deposited (Figure 3-10 A2-3). Higher magnification images reveal light staining for collagen









cellular regeneration via clonal expansion of the resident cells, mainly hepatocytes, which form

hepatic foci. They are often referred to as 'regenerative' or 'hyperplastic' foci, implying concepts

of pathogenesis rather than a morphological definition. They cannot be truly regenerative in that

restitution to normal liver tissue does not occur [80]. At the cellular and molecular level, this

progressive process is mainly characterized by cellular activation of hepatic stellate cells and

aberrant activity of TGFP and its downstream cellular mediators (CTGF). The complex signaling

pathways of this pivotal cytokine during the fibrogenic response and its connection to other

signal cascades are now understood in some detail.

Hepatic stellate cell (HSC) activation is the hallmark of fibrosis. TGFP is considered the

most powerful mediator of HSC activation in vitro and in vivo [79]. Kupffer cells (KCs) are a

main source of TGFP in the liver and promote HSC activation and fibrogenesis [1 l; 25]. Since

1989, KCs have been implicated in activating HSCs [85]. HSC activation and fibrogenesis were

almost completely suppressed in KC-depleted mice with a significant reduction of hepatic

fibrogenesis [25]. Because the transdifferentiation of quiescent HSCs to activated myofibroblast-

like HSCs is a key event in hepatic fibrogenesis [79; 86], it is likely that the sensitization of

quiescent HSCs to TGFP and KC induced activation of HSCs constitutes the main mechanism by

which inflammation promotes fibrogenesis.

HSCs become directly fibrogenic by synthesizing ECM proteins. In addition, the activated

HSC itself proliferates and amplifies the fibrogenic response. Although the precise mechanisms

responsible for HSC activation remain elusive, substantial insight is being gained into the

molecular mechanisms underlying ECM production and cell proliferation in the HSC. The

activated HSC becomes responsive to both proliferative (PDGF) and fibrogenic (TGFP)

cytokines. These cytokines activate both mitogen-activated protein kinase (MAPK) signaling,









restored. Interestingly, upon termination of CCl4 treatment, 50% of the activated HSCs

underwent apoptosis with an accompanying increase in MMPs and decrease in TIMPs [24]. In

another study by Ruchirawat acute CCl4 WAS shown to cause a blockage of DNA methylation,

and over ten years later it was shown by Varela-Morieras that hypomethylation occurs with

chronic exposure, leading to the disruption of the careful balancing act of gene regulation [1 13;

114].

Degradation of Pre-Existing Matrix

Current evidence indicates that matrix degradation is of central importance for hepatic

fibrogenesis [15; 116], but it is also involved in liver regeneration and carcinogenesis [15; 117;

118]. Matrix metalloproteinases (MMPs) are the main enzymes that degrade ECM proteins. They

have a broad substrate specificity which degrades collagens, laminins, fibronectins,

proteoglycans as well as other matrix components. Expression of MMPs in hepatocytes and

mesenchymal cells may be mediated by the transcription factor AP-1. This transcription factor is

upregulated by many pathological stimuli, including inflammatory cytokines, growth factors and

other stress signals. AP-1 represents a single component of a complex, dynamic network of

signaling pathways involved in the response to hepatic injury, so it is likely that several other

factors co-regulate MMP expression [119].

MMPs are secreted from cells into the extracellular space as proenzymes, which are

subsequently activated by HSC cell surface-associated proteases. The active enzymes are, in

turn, inhibited by a family of TIMPs. This combination of positive and negative regulation

governs matrix degradation. Additional matrix proteases have been implicated in liver ECM

remodeling. Progelatinase A is secreted by activated HSCs in culture. This proenzyme is

activated by a protein complex found within the HSC membrane. The activation of matrix

proteases at the HSC membrane indicate that the active HSC is able to independently remodel











140. Daniels JT, Schultz GS, Blalock TD, Garrett Q, Grotendorst GR, Dean NM, and Khaw
PT. Mediation of transforming growth factor-beta(1 )-stimulated matrix contraction by
fibroblasts: a role for connective tissue growth factor in contractile scarring. Am J Pathol
2003;163:2043-2052.

141. Grotendorst GR, Okochi H, Hayashi N. A novel transforming growth factor beta
response element controls the expression of the connective tissue growth factor gene.
Cell Growth Differ. 1996;7(4):469-80.

142. Chen Y, Blom IE, Sa S, Goldschmeding R, Abraham DJ, Leask A. CTGF expression in
mesangial cells: involvement of SMADs, MAP kinase, and PKC. Kidney Int.
2002;62(4):1149-59.

143. Bork P. The modular architecture of a new family of growth regulators related to
connective tissue growth factor. FEB S Lett. 1993;26:327(2): 125-3 0.

144. Kim HS, Nagalla SR, Oh Y, Wilson E, Roberts CT Jr, Rosenfeld RG. Identification of a
family of low-affinity insulin-like growth factor binding proteins (IGFBPs):
characterization of connective tissue growth factor as a member of the IGFBP
superfamily. Proc Natl Acad Sci USA. 1997;94(24): 12981-6.

145. Voorberg J, Fontijn R, Calafat J, Janssen H, van Mourik JA, Pannekoek H. Assembly and
routing of von Willebrand factor variants: the requirements for disulfide-linked
dimerization reside within the carboxy-terminal 151 amino acids. J Cell Biol.
1991;113(1):195-205.

146. Guo NH, Krutzsch HC, Negre E, Vogel T, Blake DA, Roberts DD. Heparin- and
sulfatide-binding peptides from the type I repeats of human thrombospondin promote
melanoma cell adhesion. Proc Natl Acad Sci USA. 1992;89(7):3040-4.

147. Brigstock DR. The connective tissue growth factor/cysteine-rich 61/nephroblastoma
overexpressed (CCN) family. Endocr Rev. 1999;20:189-206.

148. Pi L, Ding X, Jorgensen M, Pan JJ, Oh SH, Pintilie DG, Brown AR, Song WY, and
Petersen BE. Connective Tissue Growth Factor with a Novel Fibronectin Binding Site
Promotes Cell Adhesion and Migration During Rat Oval Cell Activation. Hepatology
epub.

149. Frazier K, Williams S, Kothapalli D, Klapper H, Grotendorst GR. Stimulation of
fibroblast cell growth, matrix production, and granulation tissue formation by connective
tissue growth factor. J Invest Dermatol. 1996; 107(3):404-11.

150. Kireeva ML, Latinkic BV, Kolesnikova TV, Chen CC, Yang GP, Abler AS, Lau LF.
Cyr6 1 and Fispl2 are both ECM-associated signaling molecules: activities, metabolism,
and localization during development. Exp Cell Res. 1997;233(1):63-77.
















































Figure 3 -12 Laminin deposition in chronic liver inj ury models by immunohi stochemi stry. (Al-
3) AA injury at dl5, 45 and 90, respectively. (Bl-3) CCl4 injury at dl5, 45 and 90,
respectively. (C1-2) NRL. (D) immunoglobulin control. (E-F) Higher magnification of
d90 AA and D90 CCl4, respectively. A and B original magnification=600X, Cl original
magnification=200X, C2-F original magnification=600X.


























kong ans


aI


Figure 1-2 Liver acinus. The three zones of the liver radiating toward the central veins. 02003
M. H. Ross All rights reserved.









Tyrowak Thenewin Pmme


Figure 1-3 Growth of remaining three liver lobes after %/ partial hepatectomy in the rat.
Compensatory hyperplasia results in the liver regaining original tissue mass in
approximately 10 to 14 days.6 01931 American Medical Association. All Rights
Reserved.


Shhuy ucullar cll






Daays aer partial hsappleoiny


Figure 1-4 Amount of resident hepatic cells within the cell cycle during the time following %/
partial hepatectomy. Hepatocytes represent the proliferative driving source behind
liver regeneration. 01997 AAAS. All Rights Reserved.


Figure 1-1 Hepatic microarchitecture and blood flow. Red arrows indicate blood flow and green
arrows indicate the direction of bile flow. 02003 M. H. Ross All rights reserved.









LIST OF ABBREVIATIONS

2AAF 2-Acetylamino fluoride

ADH Alcohol dehydrogenase

ALDH Aldehyde dehydrogenase

AA Allyl alcohol

AFP Alpha fetal protein

a-SMA Alpha-smooth muscle actin

APC Antigen presenting cell

BM Basement membrane

CCl4 Carbon tetrachloride

Coll Collagen type I

CollV Collagen type IV

CTGF Connective tissue growth factor

CYP2El Cytochrome P4502El

EC Endothelial cell

EGF Endothelial cell growth factor

EMT Epithelial-mesenchymal transition

ECM Extracellular matrix

FN Fibronectin

GFAP Glial fibrillary acidic protein

GSH Glutathione

HOC Hepatic oval cell

HSC Hepatic stellate cell

HGF Hepatocyte growth factor

IGF Insulin-like growth factor









then wrapped around the gel and affixed on a glass plate that is larger than the gel, the bubbles

were removed and it was allowed to dry overnight before it was cut to size.

Probing of Western Membrane

The membrane was blocked for 1hr at RT or overnight with a blocking solution consisting

of 5g skim milk and 100mL lX PB S, then washed with PB S-T for 5min. The membrane was

probed with the appropriate concentration of primary antibody overnight at 4oC. The membrane

was then rinsed 3x for 5min each with lX PBS. The appropriate horseradish peroxidase

conjugated secondary antibody was applied in lX PB S for 30min to 1hr shaking at RT. The

membrane was then rinsed again 3X with lX PBS.

Developing of Western Membrane with ECL PlusTM

Excess liquid was removed from the membrane and it was placed within a plastic bag.

25C1L of Solution A mixed with 1mL of Solution B ECL PlusTM reagents (GE Healthcare,

Piscataway, NJ) was incubated on the membrane for 5min. Excess ECL PlusTM reagent was

removed. Film was exposed to the membrane for 5sec to overnight depending on the brightness

of the banding pattern. The membrane was then stripped if further probing was necessary.

Membrane Stripping for Reprobing

20mL of 5X stripping solution was diluted to lX with 80ml water (100ml total). Then

714CLL P-Mercaptoethanol was added and membrane placed within the solution. The membranes

were incubated in closed Tupperware container at 56oC for no longer than 30min with

intermittent shaking. The membrane was then washed for 5-8 X 5min each with lX PB S-T until

all residual B-Mercaptoethanol was removed. Membranes were then reblocked with milk and

reprobed as mentioned.










persist. As with cirrhosis in the human liver, this matrix decreases liver function, forcing an

attempt at compensation which is evidenced by the development of regenerative nodules within

the matrix. However, these nodules lack appropriate organization and do little to resolve the

situation. Matrix accumulation in the AA model is not as extensive as in the CCl4 mOdel. Despite

this fact, chronic AA administration results in a much sicker animal than chronic CCl4. This is

likely due to continued injury by AA in the portal zone which, over time, enlarges the necrotic

area. The CCl4 induced injury does not appear to worsen significantly at later time points

suggesting that the remaining hepatocytes may not metabolize the toxin. From a clinical

standpoint, this would suggest that chronic portal zone injury may be more easily managed so

long as the source of injury is identified and removed. Central zone injury may be less damaging

to the hepatic parenchyma over time, though the resulting fibrosis is persistent and may continue

to impair liver function even after the source of injury is removed.









2. The main role in fibrosis reduction by HSCs is by their apoptosis and reduced hepatic

expression of metalloproteinase inhibitors which results in spontaneous resolution [24]. HSCs

also produce growth factors such as TGFP, connective tissue growth factor (CTGF), hepatocyte

growth factor (HGF) and epithelial growth factor (EGF) [21]. TGFP is negated by TNFu and if it

is blocked, not only are stellate cells are not activated, but the liver is protected from ethanol-

induced hepatic injury [25; 13].

HSCs may function as professional APCs for activation or restimulation of T cells through

the space of Disse when the endothelial cell barrier is injured. HSCs activate T cell responses

and express CD31 on their surface at a high density and present antigens to 1VHCI- and II-

restricted T cells for clearance [12]. According to Winau et al, HSCs (with >98.5% purity) are

capable of inducing vigorous NKT cell responses in vitro and in vivo and promoting homeostatic

proliferation of NKT cells through production of IL-15 [12].

Biliary Cells

Cells of bile ducts are commonly called biliary cells, bile duct cells or cholangiocytes. Bile

is formed and secreted by hepatocytes and flows opposite of the blood flow through the bile

caniculi into the canals of Herring and then to the bile ducts flowing into the larger common

hepatic duct. If the animal has a gal bladder it would then be stored here until needed, otherwise

it is immediately released down the common bile duct to the duodenum. If this flow is disturbed,

bile can build up into the liver and cause inflammation, fibrosis and eventually cirrhosis. Biliary

cells produce and express collagen type IV, laminin, entactin and perlecan. These cells also

produce the growth factors TGFP [26] and CTGF [27]. Oval cells reside in the same space, most

likely further down in the canal of Herring and are drawn into the liver through injury, and can

form more biliary cells or other liver cells in vivo. Biliary cells can also arise from hepatocytes,










xB, causing an increase in TNFu production. TNFu induces neutrophil infiltration and stimulates

mitochondrial oxidant production in hepatocytes, which are sensitized to undergo apoptosis [10].

KCs also release ecosinoids and leukotrine B4 which are known chemotactic factors for

neutrophils [1l]. KCs highly express F4/80 and Major Histo-Compatibility class I (MHC-I).

Expression of MHC-II, CDld, and CD86 are detected only at low amounts [12]. KCs produce

IL1, IL6, TGF-P and latent TGFP binding protein 1, 2 [13], TGFPR [14], matrix

metalloproteinases 2, 9, 13, 14 and their inhibitors, TIMP-1 [15] and TIMP-2 [16]. This triggers

stellate cells to lay down a protective layer of proteins, consisting of mainly collagen.

Pitt cells are attached to endothelial lining of the sinusoids. Pitt cells are granular cells with

a high natural killer cell activity, putatively to act against tumor and virus-infected cells [1]. Mast

cells are a rich source of nerve growth factor and other cytokines to destroy foreign materials and

accumulate in fibrotic liver. Other blood cells in the liver include neutrophils, T, natural killer

(NK) and NKT cells. Neutrophils are present in the sinusoids of the liver and migrate to the site

of inflammation and act as short-lived macrophages (1-2 days). Neutrophils and T cells express

receptors for TGFP and INFy, attracting cytokines (through chemotaxis) that are produced by

KCs, mast cells, ECs and by the neutrophils [17; 18]. Neutrophils are the first responders in an

injury; they also can activate KCs which then notify the neutrophils and other cells (i.e. stellate

cells) to release more factors.

Stellate Cells

Star-like cells were also first observed by Karl Wilhelm von Kupffer in 1876 and were

shown to reside in the space of Disse and have a nucleus of about 40um. They were termed

hepatic stellate cells (HSC) and represent 5-8% of normal liver cells [1]. These cells store

vitamin A (nearly 90% of the total amount in the body) and lipids. Since vitamin A is lipid

soluble this yields a very complimentary situation. They produce little amounts of collagen in









CHAPTER 5
FUTURE DIRECTIONS

This study would benefit from utilizing in situ hybridization for TGFP, CTGF, Col I and

IV, FN and LM. This would show where and which cells are producing these cytokines and

proteins. In addition to these already studied proteins, integrins that bind cells to the ECM can be

stained for. Also, since immune cells are so important to this model, KC, neutrophil and T cell-

specific markers would also be utilized. In addition, MMPs and TIMPs need to be studied

through IHC and in situ hybridization. It is anticipated that by understanding the molecular

mechanisms responsible for stellate cell proliferation and excess ECM production new

therapeutic targets will be identified for the treatment of liver fibrosis.

The obvious future direction would be to repeat this experiment to acquire fresh tissue to

quantify RNA and protein expression. Also, treatment would be considered and more animals

would be added to alleviate fibrosis and necrosis at varying levels of disease. Treatment could

include the addition of MMPs, siRNA, cell transplantation (oval cells and other bone marrow

cells). The use of proteinases would seem to be the most logical first step to overcome these

diseases. If this is not sufficient, interruption of the ECM proteins at the translational step would

be next. If this fails, cell transplantation therapies would be considered. The capacity of bone

marrow cells to differentiate into hepatocytes and intestinal cells was confirmed in male to

female human transplants. These cells are easier to obtain than other tissue-specific stem cells,

and more have more plasticity. This plasticity could be a problem and since oval cells in the bone

marrow could provide a more specific liver progenitor cell. Oval cells are still rather plastic, so

hepatocytes would also be tested.









receptor, low density lipoprotein receptor (LRP)-related protein/2-maCTOglobulin receptor is a

protein complex that binds CTGF and mediates internalization and degradation to negatively

regulate serum CTGF levels [166].

During liver fibrosis, CTGF expression is strongly up-regulated. This increase may be

easily measured in serum. This finding is corroborated by the fact that CTGF levels decrease in

fully developed, end stage cirrhosis, when the fibrotic process is complete [88].

Specific Aims of this Study

Chronic chemical injuries occur throughout the world as medical and industrial

environments continue to expand. The liver is responsible for the metabolism of the maj ority of

these chemicals. The resulting metabolites may either be useful or toxic by-products. Allyl

alcohol and carbon tetrachloride are chemicals which affect the general public in a variety of

ways. These chemicals injure different zones of the liver. The goal of this research was to

investigate differences in the fibrotic response to chronic hepatic portal zone injury versus

chronic central zone injury. This lead to the desire to understand how the ECM is laid down

during these chronic liver injuries. For this study we specifically asked how can protein and

nucleic acids be extracted from archival formalin Eixed paraffin embedded (FFPE) fibrotic liver?

Is a periportal injury more severe than a pericentral injury? Where are these proteins expressed,

and what is the composition of these proteins in the ECM? To begin answering these questions,

experiments were performed to:

optimize extractions from fresh, archival and chronic injury archival FFPE tissue

characterize fresh, archival and chronic injury archival FFPE tissue through IHC

This study incorporated methods from molecular genetics, molecular biology, toxicology, and

pathology.










100. Ellis EM. Reactive carbonyls and oxidative stress: potential for therapeutic intervention.
Pharmacol Ther. 2007;115(1): 13-24.

101. Morimoto K, and Takeshita T. Low Km aldehyde dehydrogenase (ALDH2)
polymorphism, alcohol drinking behavior, and chromosome alterations in peripheral
lymphocytes. Environmental Health Perspectives 1996:104;3.

102. U.S. Public Health Service, Agency for Toxic Substances: Toxicological profiles for
carbon tetrachloride. 1989.

103. Raucy JL, Kraner JC, Lasker JM. Bioactivation of halogenated hydrocarbons by
cytochrome P4502E1. Crit Rev Toxicol. 1993;23(1): 1-20.

104. Gruebele A, Zawaski K, Kaplan D, Novak RF. Cytochrome P4502E1- and cytochrome
P4502Bl1/2B2-catalyzed carbon tetrachloride metabolism: effects on signal transduction
as demonstrated by altered immediate-early (c-Fos and c-Jun) gene expression and
nuclear AP-1 and NF-kappa B transcription factor levels. Drug Metab Dispos.
1996;24(1):15-22.

105. Gomez MI, Castro JA. Reaction of trichloromethyl free radicals with deoxyribonucleic
acid bases. Res Commun Chem Pathol Pharmacol. 1981;32(1):147-53.

106. Connor HD, Lacagnin LB, Knecht KT, Thurman RG, Mason RP. Reaction of glutathione
with a free radical metabolite of carbon tetrachloride. Mol Pharmacol. 1990;37(3):443-
51.

107. Rechnagel RO and Glende EA. Carbon tetrachloride hepatotoxicity: an example of lethal
cleavage. CRC Crit Rev Tox. 1973;2:263-297.

108. Farber E and El-Mofty SK. The biochemical pathology of liver cell necrosis. Am J
Pathol. 1975;81:237-250.

109. Basu S. Carbon tetrachloride-induced lipid peroxidation: eicosanoid formation and their
regulation by antioxidant nutrients. Toxicology 2003;189(1-2): 113-27.

110. Beddowes EJ, Faux S and Chipman K. Chloroform, carbon tetrachloride and glutathione
depletion induce secondary genotoxicity in liver cells via oxidative stress. Toxicology
2003;187:101-115.

111. Smuckler E. Structural and functional changes in acute liver injury. Environmental
Health Perspectives 1976; 15:13-25.

112. Weber L, Boll M, Stampfl A. Hepatotoxicity and Mechanism of Action of Haloalkanes:
Carbon Tetrachloride as a Toxicological Model. Critical Reviews in Toxicology
2003;33(2):105-136.









Acetone Precipitation

The appropriate amount of acetone was cooled to -200C and added in four times the

volume of the protein/RIPA buffer solution. The samples were either vortexed or inverted then

incubated at -200C for 1hr to overnight, and then were centrifuged for 10min at 14000xg at 40C.

The supernatant was saved for further extraction. The pellet was allowed to dry at RT for 10-

25min, not allowing it to over dry. Buffer is added to reconstitute the protein in the appropriate

amounts depending on the pellet size and subsequent spectrophotometer reading (DC protein

assay). This step was repeated as necessary.

Protein Quantification with DC Protein Assay

Blank and protein standards were made in 1ml tube with luL sample and 24CLL RIPA

Buffer without Protease Inhibitor Solution in 1ml clear tube. In another 1ml tube 125C1L Reagent

A per reaction and 2.5C1L Reagent S per reaction from the DC Protein Assay (Bio-Rad) were

mixed. Note: Reaction number equals Sample number plus 5. 125C1L of combined solutions A&

S was added to each reaction. When ready to measure Iml Reagent B was added and tubes

vortexed. 5-10min after the addition of solution B the OD of the samples were measured in

disposable cuvettes in Spectrophotometer set to 750nm.

Running Gel

The sponge on gel pouring apparatus was dampened and the plates cleaned with alcohol.

The plates were then aligned and secured in gel pouring apparatus. The gels were poured using

electronic or plastic Pasteur pipettes. They were poured to approximately 1-1.25cm below the

top of the plate. Butyl alcohol was added to top of glass to prevent bubbles and smiling of gel.

After 30min the remaining solution in the 15ml tube was inverted to determine if gel solidified.

The butyl alcohol was removed and the gel rinsed with Milli-Q H20 until the odor dissipated.









the initial formalin Eixation, was hampering migration through the gel. Acidiaication or alkylation

is a well known method for disrupting formalin crosslinking of proteins for

immunohi stochemi stry. Samples were treated with buffers of various pH with or without a cation

chelators. However, proteins still could not be reproducibly resolved by electrophoresis (Figure

3-1 B).

Attempts to isolate amplifiable DNA from archival tissue met with marginal success

(Figure 3-1 C, D). Faint bands of amplified DNA were identifiable following PCR amplification.

The isolation procedure proved to be unreliable in terms of reproducibility. RT-PCR of RNA

from archival tissue seemed very unlikely, so we chose to concentrate on immunohistochemistry

to describe the fibrotic response to chronic liver injury.

Immunohistochemical Analysis of Archival Tissue

Analysis of H&E stained tissues gave the expected results (Figure 3-2). Chronic AA

induced periportal necrosis and CCl4 induced pericentral necrosis. We were intrigued that early

time points form the AA treated animals demonstrated a very mild initial injury as compared to

the CCl4 mOdel. This was somewhat surprising as these animals appeared to be much sicker

throughout the course of treatment. Later time points provided an answer to this contradiction.

While the AA injury induced a minor initial injury, damage continued to accumulate over the

course of treatment. In contrast to this, the CCl4 mOdel induced relatively severe injury in the

early to mid time points (Figure 3-2). However, as treatment continued, the rate of tissue

destruction appeared to slow. A possible explanation for this lies in the toxicology underlying

each of these injury models.

AA is metabolized by the portal zone cells which may be considered relatively immature

and have a low ploidy of only 2N. Oval cells that proliferate from the portal zone subsequently

differentiate into immature hepatocytes that are able to metabolize AA. These cells would be









14. Kossmann T, Manthey CL, Brandes ME, Morganti-Kossmann MC, Ohura K, Allen JB,
Mergenhagen SE, Wahl SM. Kupffer cells express type I TGF-beta receptors, migrate to
TGF-beta and participate in streptococcal cell wall induced hepatic granuloma formation.
Growth Factors 1992;7(1):73-83.

15. Iredale JR, Benyon RC, Arthur MJ, Ferris WE, Alcolado R, Winwood PJ, et al. Tissue
inhibitor of metalloproteinase-I messenger RNA expression is enhanced relative to
interstitial collagenase messenger RNA in experimental liver injury and fibrosis.
Hepatology 1996;24:176-84.

16. Knittel T, Mehde M, Kobold D, Saile B, Dinter C, Ramadori G. Expression patterns of
matrix metalloproteinases and their inhibitors in parenchymal and non-parenchymal cells
of rat liver: regulation by TNF-alpha and TGF-betal. J Hepatol. 1999;30(1):48-60.

17. Brandes ME, Mai UE, Ohura K, Wahl SM. Type I transforming growth factor-beta
receptors on neutrophils mediate chemotaxis to transforming growth factor-beta. J
Immunol. 1991; 147(5):1600-6.

18. Grotendorst GR, Smale G, Pencev D. Production of transforming growth factor beta by
human peripheral blood monocytes and neutrophils. J Cell Physiol. 1989;140(2): 396-402.

19. Liver Growth and Repair. Strain A, Diehl AM (eds.) New York. Chapman and Hall
(1998).

20. Neubauer K, Knittel T, Aurisch S, Fellmer P, Ramadori G. Glial fibrillary acidic protein--
a cell type specific marker for Ito cells in vivo and in vitro. J Hepatol. 1996;24(6):719-30.

21. Geerts A. History, heterogeneity, developmental biology, and functions of quiescent
hepatic stellate cells. Semin Liver Dis. 2001;21(3):311-35.

22. Wahl SM, Allen JB, Weeks BS, Wong HL, Klotman PE. Transforming growth factor
beta enhances integrin expression and type IV collagenase secretion in human
monocytes. Proc. Na. Ac. Sci. USA 1993;90:4577-458.

23. Balazovich KJ, Fernandez R, Hinkovska-Galcheva V, Suchard SJ, Boxer LA.
Transforming growth factor-betal stimulates degranulation and oxidant release by
adherent human neutrophils. J Leukoc Biol. 1996;60(6):772-7.

24. Arthur MJP. Fibrogenesis II. Metalloproteinases and their inhibitors in liver fibrosis. Am
J Physiol Gastrointest Liver Physiol 2000;279:245-249.

25. Wheeler MD, et al. The role of Kupffer cell oxidant production in early ethanol-induced
liver disease. Free Radical Biology & Medicine 2001;31:12: 1544-1549.


















H AA CCl4 P-Value
dl5 7+/- 3 14+/-4 0.0003
d30 2+/-1 7+/-2 0.0001
d45 9+/-5 36+/-14 0.0001
d60 7+/-4 8+/-3 0.5350
d70 6+/-10 30+/-7 0.0001
d90 14+/-14 28+/-16 0.0518


NH AA CCl4 P-Value
dl5 6+/-2 30+/-5 0.0001
d30 10+/-3 15+/-7 0.0525
d45 25+/-12 9+/-3 0.0007
d60 28+/-10 27+/-6 0.7894
d70 31+/-18 15+/-5 0.0144
d90 91+/-61 20+/-4 0.0017


HSCs AA CCl4 P-value

dl5 1+/-2 27+/-8 0.0001

d45 27+/-7 74+/-20 0.0011

d90 59+/-11 82+/-17 0.0347


Table 3-1 Proliferating (BrdU ) hepatocytes (H) and non-hepatocytes (NH) over the time course
of chronic AA and CC14l41VeT injurieS. Column one of each H and NH represents the day
of treatment, column two and three are the average number of cells positive for BrdU
after AA and CCl4 treatment, and column four shows the statistical difference between
the two chronic injury models represented by a p-value generated from t-test analysis for
each time point.


Table 3-2 Activation of HSCs over the time course of chronic AA and CCl4. COlumn one shows
the days of treatment. Column two represents the average number of HSCs in a 200x
in field for AA. Column three represents the average number of HSCs in a 200x field
for CCl4. COlumn four represents the p-value determined by a t-test between the two
models during the time course.









effect upon the oval cells, which may have resulted in the inhibition of their activation. This

study does support the idea that oval cells arise from the periportal zone, specifically from the

canals of Hering [3 5]. Petersen reported in 1999 that oval cells also come directly from the bone

marrow [47].

Liver Regeneration

Under normal conditions, hepatocytes exhibit nominal replicative activity; only 1 in every

20,000 hepatocytes undergoes mitotic division at any one given time point. Yet, hepatocyte

division is the maj or driving force behind liver regeneration [48]. Compensatory hyperplasia of

the liver (Figure 1-4), takes place in response to mild to severe liver injury resulting from

surgical resection of a portion of the liver or exposure to destructive agents such as hepato-toxins

or hepatotropic viruses. This term is utilized instead of regeneration due to the unusual response

to gain enough tissue to function normally. Instead of gaining individual lobules back it will

form one large lobule. The liver must be a certain size to do all the jobs necessary to keep at

homeostasis. In 1931 Higgins and Anderson reported the first partial hepatectomy (PH),

removing 69% of the liver in a rat [49]. Removal of as much as 80-90% of the liver can be

restored in the absence of disease [1]. Rats have been subj ected to 2/3 hepatectomy (12 times in

all) with the liver able to regenerate itself. Each time, the liver is restored to its normal size

within a few weeks (Figure 1-3). It has been estimated that one rat hepatocyte has the capacity to

generate at least 50 livers [30].

In the rat, hepatocytes move from the GO resting phase of the cell cycle into Gl, as

mediated by the cyclin Dl pathway within 15hrs of PH. Periportal hepatocytes are the first to

undergo DNA synthesis and proliferation gradually spreads to include the hepatocytes located

around the central vein. A large peak of DNA synthesis is observed at about 24hrs post PH, and a

second, yet smaller peak at 48hrs. The smaller peak reflects DNA synthesis occurring in non-









38. Nagy P, Kiss A, Schnur J, Thorgeirsson SS. Dexamethasone inhibits the proliferation of
hepatocytes and oval cells but not bile duct cells in rat liver. Hepatology 1998;28(2):423-


39. Omori M, Evarts RP, Omori N, Hu Z, Marsden ER, Thorgeirsson SS. Expression of
alpha-fetoprotein and stem cell factor/c-kit system in bile duct ligated young rats.
Hepatology 1997;25(5):1115-22.

40. Omori N, Omori M, Evarts RP, Teramoto T, Miller MJ, Hoang TN, Thorgeirsson SS.
Partial cloning of rat CD34 cDNA and expression during stem cell-dependent liver
regeneration in the adult rat. Hepatology 1997;26(3):720-7.

41. Fujio K, Evarts RP, Hu Z, Marsden ER, Thorgeirsson SS. Expression of stem cell factor
and its receptor, c-kit, during liver regeneration from putative stem cells in adult rat. Lab
Invest. 1994;70(4):511-6.

42. Golding M, Sarraf CE, Lalani EN, Anilkumar TV, Edwards RJ, Nagy P, Thorgeirsson
SS, Alison MR. Oval cell differentiation into hepatocytes in the acetylamino-fluorene
(2AAF) treated regenerating rat liver. Hepatology 1995;22:1243-1253.

43. Evarts RP, Nagy P, Nakatsukas H, Marsden E, Thorgeirsson SS. In vivo differentiation of
rat liver oval cells into hepatocytes. Cancer Res. 1989;49:1541.

44. Fausto N, Campbell JS. Mech The role of hepatocytes and oval cells in liver regeneration
and repopulation. Dev. 2003;120(1): 117-30.

45. Nagy P, Bisgaard HC, Santoni-Rugiu E, Thorgeirsson SS. In vivo infusion of growth
factors enhances the mitogenic response of rat hepatic ductal (oval) cells after
administration of 2-acetylaminofluorene. Hepatology 1996;23(1):71-9.

46. Hasuike et al. Hepatocyte growth factor accelerates the proliferation of hepatic oval cells
and possibly promotes the differentiation in a 2-acetylaminofluorene/partial hepatectomy
model in rats. Journal of Gastroenterology and Hepatology 2005;20: 1753-1761.

47. Petersen BE, et al. Bone Marrow as a Potential Source of Hepatic Oval Cells. Science
1999;284.

48. Martinez-Hernandez A, Amenta PS. The extracellular matrix in hepatic regeneration.
FASEB.1995;9.

49. Higgins GM and Anderson RM. Experimental pathology of the liver:I. Restoration of the
liver of the white rat following partial surgical removal. Arch Pathol. 193 1;12: 186-202.









Optimizing Protein and DNA Extractions from Archival Formalin Fixed Paraffin
Embedded Tissue

Despite repeated attempts at purification, electrophoretic resolution of discrete protein

bands from archival tissue was not possible (Figure 3-1B). However, we attempted to resolve

beta actin on Western blots of archival tissue. Fresh NRL displayed a strong band of the

appropriate molecular weight (Figure 3-1D). There was no corresponding band in the archival

fibrotic tissue lanes (Figure 3-1D). Interestingly, fresh formalin Eixed NRL also failed to produce

a beta actin band by Western blot (Figure SA).

Amplifieation of a region of the GAPDH gene in DNA isolated from archival, paraffin

embedded NRL was possible (Figure 3-1C). However, PCR amplification of DNA isolated from

archival, paraffin embedded AA and CCl4 treated liver was less successful (Figure 3-1D). As

double stranded DNA is much more stable than single stranded RNA, RT-PCR was not

attempted on archival tissue.

Overall Evaluation of the Two Injury Models

As expected, the AA model showed signs of necrosis around the portal triad (PT) with

relatively little damage around the central vein (CV) (Figure 3-2 A6). In addition, ductular

proliferation, connective tissue accumulation and cirrhosis were evident in this model.

Accumulation of small cells with morphology consistent with oval cells was also seen in this

model. In contrast to the AA model, CCl4 injury induced mainly centrilobular necrosis. Also

apparent in this model were ductular proliferation, micro nodule formation, lipid storage within

hepatocytes (steatosis) and massive fibrosis eventually leading to cirrhosis.









Fibrosis is a well-known histological and biochemical hallmark of cirrhosis, but fibrosis does not

necessarily accompany cirrhosis. Originally fibrosis was defined by a WHO expert group in

1978 "as the presence of excess collagen due to new fiber formation" [80]. Early clinical reports

in the 1970s suggested that advanced liver fibrosis is potentially reversible. Even with this

known, liver fibrosis received little attention until the 1980s, when HSCs were identified as the

main collagen-producing cells in the liver [81].

The accumulation of ECM proteins alters the hepatic architecture by forming scar tissue.

The subsequent development of nodules of regenerating hepatocytes defines cirrhosis. The

nodules of a cirrhotic liver lack normal lobular organization and are surrounded by fibrous tissue

which is formed by a combination of necrotic cellular debris and de novo ECM protein. Cirrhosis

produces hepatocellular dysfunction and increased intrahepatic resistance to blood flow, which

result in hepatic insufficiency and portal hypertension, respectively. In cirrhosis, increased

pressure in the portal vein causes large veins (varices) to develop across the esophagus and

stomach to bypass the blockage. The varices become fragile and can bleed easily. The essential

features of cirrhosis are considered to be parenchymal necrosis, regeneration, and diffuse

fibrosis, resulting in disorganization of the lobular architecture throughout the whole of the liver.

Toxic injury to the liver causes cellular death through both apoptosis and necrosis. Any

cellular injury results in sustained elevation of Ca2+ Signaling, which triggers necrotic or

apoptotic cellular death [82]. Particularly important is an activation of nuclear factor kappa B

(NF-icB), which regulates production of cytokines and interferons [83]. Prolonged stress to the

cell, particularly the endoplasmic reticulum, can stimulate pro-caspase-12, localized in the

endoplasmic reticulum. When activated, caspase 12 stimulates other pro-apoptotic caspases [84].

After cell death, new cells must come in to replace the dead ones. Chronic injury results in










possible that as mature hepatocytes enter the cell cycle, they become transiently able to

metabolize AA and are either destroyed or their proliferative ability is diminished. Since the

periportal tissue is damaged, oxygen and nutrients are not able to reach the rest of the liver by the

later time points leading to additional stress and necrosis. This massive necrosis that develops in

these animals over time recruits inflammatory cells. There is robust proliferation of non-

parenchymal cells in the AA model, particularly at the later time points. We suspect that these

inflammatory cells are responsible for the high rate of proliferation seen in the non-parenchymal

fraction.

In contrast to our AA chronic injury model, CCl4 is metabolized by mature central zone

hepatocytes with an average of 4N. Destruction of these cells would result in the proliferation of

oval cells and portal zone hepatocytes that would, for a time, be unable to metabolize the toxin.

The periportal cells, as mentions above are on average only 2N and can rapidly divide to

repopulate the liver. Proliferation data shows that the CCl4 mOdel induces robust proliferation of

hepatocytes by d45 followed by a lack of proliferating cells by d60 then a new surge of

hepatocyte proliferation is seen by d90 within the regenerating nodules (Figure 3-3). This is the

expected regenerative response to the loss of hepatic tissue. There is modest proliferation of the

non-parenchymal cell fraction in this model through all time points. These cells likely contain

inflammatory cells recruited to the necrotic areas, as well as fibroblasts and oval cells that are

participating in tissue regeneration and remodeling. Because the periportal tissue is not

completely damaged, CCl4 injured liver can still receive nutrients and oxygen, except perhaps to

the hepatic foci which were created by the response to the reactive oxygen species created from

the metabolism of this chemical as seen by Jung for AA and by the numerous studies of chronic

CCl4 treatment. These walled off foci might be called regenerating nodules in this case because,










average and standard deviations and the p-values were found using the t-test from

http://www.graphpad. com. The significance between AA and CCl4 at days 15, 30, 45, 60, 70 and

90 for each hepatocyte and non-hepatocytes for BrdU and for stellate cells, days 15, 45 and 90 of

treatment for a-SMA were in put into the t-test.

Table 2-1 List of antibodies utilized
Primary Cat # Co. Dilution Retrieval Secondary
BrdU MO744 Dako 1:200 None Donkey anti-mouse
Fibronectin ab23751 Abcam 1:200 Trypsin Digest Donkey anti-mouse
CTGF ab6992 Abcam 1:800 None Goat anti-mouse
Collagen IV abl3966 Abcam 1:100 Trypsin Digest Goat anti-mouse
a-SMA A5228 Sigma 1:200 Trypsin Digest Donkey anti-mouse
Laminin 20097 Dako 1:25 Proteinase K Goat anti-rabbit
TGF[3 sc-146 Santa Cruz 1:100 Citra Goat anti-rabbit
AFP A0008 Dako 1:600 Trlg Goat anti-rabbit
CD31 sc-1506 Santa Cruz 1:50 Citra Rabbit anti-goat
vWF sc-8068 Santa Cruz 1:600 Citra Rabbit anti-goat









minutes then in lX transfer buffer for 20min. Sponges and filter papers were also soaked in

transfer buffer. The gel plates were opened and the stacking gel/wells were removed. The gel

was submerged in lX transfer buffer. A sandwich consisting of black assembly tray, sponge,

filter paper, gel, PVDF membrane, filter paper, sponge, and red assembly tray along with an ice

block and stir bar were placed in the transfer apparatus. The transfer apparatus was filled with

lX transfer buffer and placed on a stir plate. The transfer buffer was stirred continuously while

transferring to ensure the apparatus would not overheat. The proteins were transferred at 150

milliamps for 60min for a 75mm gel and 90min for a 150mm gel.

Coomassie Staining of Western Gel

In order to determine protein banding patterns or efficiency of transfers, the gel was placed

in 20mL of methanol, 7.5mL of glacial acetic acid, and 72.5mL of Milli-Q H20 and shaken at

RT for 30min. The gel was stained with Coomassie (0.4g of Coomassie blue R350 (Sigma),

80mL methanol, 40mL glacial acetic acid, and 280mL Milli-Q H20) for 30Omin at RT. It was

then destined as desired in 500ml methanol, 100ml glacial acetic acid, and 400mL Milli-Q H20.

A Kimwipe@ (Kimberly-Clarke, Roswell, GA) was placed in the solution and as it was saturated

with color it was replaced until the desired level of destaining was achieved.

Silver Staining of Western Gel

The SilverQuestTM Silver Staining Kit (Invitrogen, Carlsbad, CA) provides a more

sensitive detection of proteins than the Coomassie stain. All solutions were made at the time of

use. First the gel was soaked in 7% acetic acid for 7min, transferred to 200mL of 50% methanol

for 20min X 2, rinsed in 200mL water for 10min X 2, and 5min before the end of the final water

rinse the staining solution was made. The gel was then soaked in the staining solution for 15min,

rinsed in 200mL of water 5min X 2. The gel was soaked in developing solution until bands were

visible, about 2 to 15Smin then stopped by rinsing in water 3 changes of water. Cellophane was









interactions of adj acent molecules, both laterally and at either terminus via disulfide bonding.

Free deposits of CollV are found in the subsinusoidal space to increase tensile strength to make

up for the lack of basement membrane around the space of Disse [2003, Yang]. It is found

throughout the liver, in both the central and periportal zones. Within the ECM, TGFP can bind

CollV [48].

Fibronectin

In normal liver Fibronectin (FN) is found in the subcapsular connective tissue, septa, portal

areas surrounding interstitial cells, and the space of Disse. FN is a glycoprotein that has functions

associated with cell adhesion and migration. FN organizes cell-cell interactions and cell-ECM

contacts by binding to different components of the ECM and to membrane-bound FN receptors

(integrins) on cell surfaces [70]. It is found in BMs and throughout the ECM. It exists in two

forms, termed cellular FN and plasma FN [48]. Cellular FN is made by fibroblasts, chondrocytes,

endothelial cells, macrophages, as well as some epithelial cells. In the liver, HSCs and KCs are

responsible for its production and deposition. FN is deposited in the ECM as filaments which are

significantly insoluble. Plasma FN is synthesized by hepatocytes and represents about 1% of

serum protein which is about 300 micrograms per mL [48]. The composition of FN depends on

the tissue type and/or cellular conditions.

FN is made up of about 5% carbohydrates which bind to proteins such as integrins,

collagens, fibrin, heparin and it also binds to CTGF. FN anchors cells to collagens for adhesion

[48]. It also alters cell morphology and surface architecture. The main function of FN is its

involvement in cellular migration during development and regeneration, and regulation of cell

growth and differentiation [70]. According to Yang et. al, FN can serve as a chemoattractant

stimulus for HSCs [65]. FN also is a part of the provisional matrix during oval cell mediated









neuron survival, and regulation of erythroid differentiation [64]. HGF is found in large amounts

in the ECM where it can create a local environment and be copiously released after injury. HGF

may inhibit TGFP expression [19]. EGF leads to a spread of HSCs within the space of Disse,

perpetuating fibrotic disease [65]. A combination of TGFP and EGF induce HSCs to migrate

[65].

The space of Disse has a low-density ECM that allows rapid bidirectional macromolecular

exchange with plasma. Around vessels, the ECM becomes denser, as it is packed with fibrillar

collagens (except around bile ducts), structural collagens, laminins, heparin sulfate

proteoglycans, fibronectins and integrins. Hepatic ECM separating parenchyma from sinusoids

in the space of Disse is characterized by a basement membrane-like matrix with a low density

and is critical for maintaining the differentiated functions of resident hepatic cells.

Alpha-Smooth Muscle Actin

Actin makes up the most abundant protein in many eukaryotic cell types. It polymerizes

forming microfilaments that have an array of functions including regulating contractility,

motility, cytokinesis, phagocytosis, adhesion, cell morphology, and providing structural support.

It binds the intracellular portions of integrins to aid in cell attachment to the ECM. Alpha-

Smooth muscle actin (a-SMA), an isoform typical of smooth muscle cells (SMC) and present in

high amounts in vascular SMC, was demonstrated in the cytoplasm of pericytes of various rat

and human. In SMC and pericytes, a-SMA is localized in microfilament bundles for contractile

functioning [66]. The first 125bp of the a-SMA promoter contains three regions that allow

TGFP-induced activation. A mutation in any one of these elements completely abolishes

transcriptional activity [67]. HSCs (pericytes) produce a-SMA which is deposited in the

regenerating and damaged areas. This contractile protein supports cell motility and acts as a

structural support for proliferating cells.










113. Ruchirawat M, Mostafa MH, Shank RC and Weisburger EK. Inhibitory effects of carbon
tetrachloride on diethylnitrosamine metabolism and DNA alkylation. Carcinogenesis
1983;4:537-539.

114. Varela-Moreiras G, Alonso-Aperte E, Rubio M, Gasso M, Deulofem R, Alvarez L,
Caballeria J, Rodes J and Mato JM. Carbon tetrachloride-induced hepatic injury is
associated with global DNA hypomethylation. Hepatology 1995;22: 1310-3115.

115. Iredale JR Matrix turnover in fibrogenesi s. Hepato-Gastroenterol .1996; 43:5 6-7 1.

116. Arthur MJ. Role of Ito cells in the degradation of matrix in liver. J Gastroenterol Hepatol
1995;10:57-62.

117. Arii S, Mise M, Harada T, Furutani M, Ishigami S, Niwano M, et al. Overexpression of
matrix metalloproteinase 9 gene in hepatocellular carcinoma with invasive potential.
Hepatology 1996;24:316-22.

118. Ashida K, Nakatsukasa H, Higashi T, Ohguchi S, Hino N, Nouso K, et al. Cellular
distribution of 92-kd type IV collagenase/gelatinase B in human hepatocellular
carcinoma. Am J Pathol. 1996;149:1803-11.

119. Hess J, Angel P and Schorpp-Kistner M. AP-1 subunits: quarrel and harmony among
siblings. Journal of Cell Science 2004; 11 7:5965 -5973.

120. Gauldie J, Bonniaud P, Sime P, Ask K, Kolb M. TGF-beta, Smad3 and the process of
progressive fibrosis. Biochem Soc Trans. Aug 2007;35(4):661-4.

121. Burt DW, Law AS. Evolution of the transforming growth factor-beta superfamily. Prog
Growth Factor Res. 1992;5:99-118.

122. Roberts AB, Frolik CA, Anzano MA, Sporn MB. Transforming growth factors from
neoplastic and nonneoplastic tissues. Fed Proc. 1983;42(9):2621-6.

123. Brenner DA, Chojkier M. Acetaldehyde increases collagen gene transcription in cultured
human fibroblasts. J Biol Chem. 1987;262(36):17690-5.

124. Daley SR, Ma J, Adams E, Cobbold SP, Waldmann H. A key role for TGF-beta signaling
to T cells in the long-term acceptance of allografts. J Immunol. 2007; 179(6):3648-54.

125. Ignotz RA, Endo T, Massague J. Regulation of fibronectin and type I collagen mRNA
levels by transforming growth factor-beta. J Biol Chem. 1987;262(14):6443 -6.

126. Carr BI, Hayashi I, Branum EL, Moses HL. Inhibition of DNA synthesis in rat
hepatocytes by platelet-derived type beta transforming growth factor. Cancer Res.
1986;46(5):2330-4.









Basement Membranes in the Liver

Basement membranes (BM) are considered to be a part of the ECM. Nerves, bile ducts,

arterial, and venous vessels have BM. In contrast, sinusoids are devoid of BM. Although

perlecan and some Col IV are found in the space of Disse, neither laminin nor entactin is present,

suggesting that at least one of these components is needed for the assembly of a BM. In other

glandular structures, such as the thyroid and pancreas, endothelial and the epithelial BMs exit.

The hepatocyte is the only epithelial cell in the body not separated from the vascular space by

two continuous BMs [1]. BMs allow a rapid bidirectional macromolecular exchange between

plasma and hepatocytes, but are not as tolerant as the rest of the ECM. The formation of BM-like

structures in the space of Disse is characteristic of cirrhosis, a marked decrease in

macromolecular exchange which leads to cell starvation, toxin buildup, and ultimately cell death

[48]. BM matrix integrity, composition, and cell-matrix interactions play an important role in

anchoring HSCs and preventing them from spreading within the space of Disse and potentially

elsewhere in the liver [65].

Introduction to Liver Fibrosis

Hepatic fibrosis is a scarring process that is associated with an increased and altered

deposition of extracellular matrix in the liver [78]. It was historically thought to be a passive and

irreversible process due to the collapse of the hepatic parenchyma leading to septa-forming

condensation of pre-existing stroma. However, liver fibrosis is an active wound-healing process.

Generally, renewing cells are more vulnerable to chemical injury than intermediate cells, which

are largely quiescent.

The main causes of liver fibrosis include chronic HCV infection, alcohol abuse, and

nonalcoholic steatohepatitis (NASH). Chronic hepatic inflammation is tightly linked to fibrosis

in virtually all individuals with liver disease and in experimental models of fibrogenesis [79].

















































Figure 3-7 CTGF expression during chronic liver injury. (Al-3) AA injury at dl5, 45 and 90,
respectively. (Bl-3) CCl4 injury at dl5, 45 and 90, respectively. (C1-2) NRL. (D)
immunoglobulin control. (E-F) Higher magnification of d90 AA and D90 CCl4,
respectively. A and B original magnification=600X, Cl original magnification=200X,
C2-D original magnification=600X, E-F original magnification=200X.









between oval cells and stellate cells may form the structural basis for the cross talk between these

two cell types [60].

Overview of the Extracellular Matrix

Most of the cells in multicellular organisms are surrounded by a complex mixture of

nonliving material that makes up the extracellular matrix (ECM). Hepatic ECM that separates

parenchyma from sinusoids in the space of Disse is characterized by a basement membrane-like

matrix with a low density and is critical for maintaining the differentiated functions of resident

hepatic cells. The structure of the ECM in normal and fibrotic liver is based mainly upon

collagens, proteoglycans and glycoproteins. This matrix is altered rapidly to the slightest

metabolic changes of the cells producing it and then it in turn alters the biochemical and

morphologic phenotype of those cells [61]. It can act as a positive as well as a negative regulator

of functional differentiation depending on the cell type and the genes studied [62]. The ECM, a

modulator of many cellular functions, controls and maintains the internal environment of all

functions and structures while dispensing energy in the immediate distant area.

The matrix binds growth factors and other signaling proteins to itself to create

microenvironments [61; 19]. Cells attach to the ECM by means of transmembrane glycoproteins

called integrins. The extracellular portion of integrins binds to various types of ECM proteins,

the maj ority of which is made up from collagens, laminins and fibronectin. The intracellular

portion binds to the actin filaments of the cytoskeleton. The ECM components may function as

an affinity matrix for binding and immobilizing soluble growth and differentiation factors such

as TGFP, which binds to collagen IV, and control local regulation of these functions [63]. HGF

and EGF are also maj or factors in this matrix.

HGF is involved in the induction of cell proliferation and motility, induction of

morphogenesis, stimulation of T cell adhesion to endothelium and migration, enhancement of









normal liver [19]. They are also defined by expression of ectodermal neural markers such as glial

fibrillary acidic protein (GFAP) [20].

HSC numbers increase significantly upon liver damage [21]. Liver fibrosis is a common

consequence of chronic liver diseases and is an indirect result of the activation of HSCs. After

liver damage, HSCs undergo a transition from a quiescent to an activated phenotype. The storage

of fat or vitamin A dwindles, leading to vitamin A deficiencies and associated disorders.

Activated HSC express a-smooth muscle actin (a-SMA), and multiple chemokines and growth

factor receptors, including, leptin, endothelin and platelet-derived growth factor beta (PDGF)

[21]. Activated HSCs upregulate the expression of TGFP and establish an autocrine-loop

whereby this cytokine upregulates the expression of collagen type I genes by transcriptional

mechanisms involving peroxides [22; 23]. Both smooth muscle associated proteins are only

expressed on HSCs when they are activated through liver injury. This is for contractile

capabilities and regulating portal blood flow. Although HSCs may be the most important source

of myofibroblasts in injury, HSCs are not the sole source. The portal area of the normal liver

contains fibroblast-like cells (myofibroblasts), and the proximal bile ductules, up to the canals of

Hering, are encircled by smooth muscle cells, which may proliferate, migrate and contribute to

ECM production in response to biliary injury.

In addition to actin, activated HSCs produce and deposit structural proteins such as

collagen types I, III and IV, tenascin, indulin, laminin, entactin and perlecan directly into the

space of Disse [21]. These proteins are deposited in areas of inflammation in an effort to contain

the spread of infection. Until the 1980s, hepatocytes where considered the main source of

extracellular matrix, but in fact HSCs are the main culprits. However, they can also produce

proteinases, MMPs that degrade the ECM all while also producing their inhibitors TIMP-1 and -









CTGF was first identified in conditioned media containing human umbilical vein

endothelial cells in culture and was found to be mitogenic and chemotactic to cells of connective

tissues such as fibroblasts [156]. Besides fibroblasts, it is expressed in, and acts on endothelial

cells, skeletal and smooth muscle cells, chondrocytes, epithelial cells, neural cells and

hepatocytes. It has also been implicated in several normal physiological processes, including

those related to embryo development and differentiation [150], endochondral ossification [157],

and female reproductive tract function in the uterus and ovary [158]. Upregulation of CTGF has

been linked to many pathogeneses including fibrosis [159], tumor desmoplasia [160], wound

healing, and tissue regeneration. A rodent model with CTGF knocked out showed impaired

chondrocyte proliferation, angiogenesis, ECM production and turnover and abnormal skeletal

growth [161].

In liver, CTGF is normally expressed in a very low level, except for a local upregulation in

portal and central vein endothelia and in myocytes of portal arteries [27]. However, a significant

induction of CTGF transcripts has been found in liver injury [162] and pathogenesis including

liver fibrosis in animal models and in human chronic liver diseases [Paradis, 1999; 164].

Activated stellate cells are found to be a major cellular source of CTGF in fibrotic and CCl4-

injuried rat livers. When added in primary stellate cell cultures, CTGF strongly promote stellate

cell proliferation, migration and produce collagen type I [163]. Recently, proliferating epithelial

cells in bile duct and in hepatocytes have also been shown to produce CTGF mRNA by studying

the temporospatial expression ofCTGF in rats with acute and chronic hepatic fibrogenesis [27;

88] CTGF is known to induce phosphorylation of p42/44 MAPK and protein kinase B in primary

mesangial cells via a P integrin dependent manner [165]. These signaling pathways are likely the

mechanism through which CTGF induces phenotype changes in target cells. The multiligand









protein thiols alters secondary and tertiary protein structure leading to conformational changes

and loss of enzymatic activity [95; 98].

To date, very little data exists on the chronic exposure of AA. Jung reported that after 16

weeks of twice weekly exposure (0.62 mmol/kg) severe fibrosis occurs [94].

Chronic Carbon Tetrachloride Exposure

Carbon tetrachloride (CCl4) does not occur naturally. It is produced by chlorination of

carbon disulfide. It may be a byproduct of dichloromethane and chloroform production as it was

first discovered in 1839 by H.V. Regnault, a French chemist and physician. CCl4 is an ozone

depleting gas, and has a strong aromatic odor. CCl4 is Very stable in air; evaporates quickly and

has a half-life of 30-100 years [102]. It has been shown not to bio-accumulate in animals or soil.

It has been used commercially in the production of pesticides, solvents, and antiseptics.

CCl4 is primarily metabolized by the membrane bound enzyme cytochrome P4502El

(CYP2El), which mediates reduction of the toxin [103; 1]. CYP2B 1 and B2 are also capable of

this action with this chemical [104]. CYP2El is found more prominently on the membranes of

the endoplasmic reticulum (ER) [1]. CYP2El is constitutively expressed at low levels

throughout the body with very high levels of expression in liver, and to a lesser extent in skin,

lung and intestine. In the liver, CYP2El is concentrated within the centrilobular region,

specifically in the first 3-4 layers of hepatocytes around the central vein [1]. Here, CCl4 is

transformed into a trichloromethyl free radical [1]. hr vitro studies indicated that this radical may

interact directly with all four DNA bases, with a particular affinity for guanine and adenine

[105].

Trichloromethyl free radicals may also react immediately with lipids and cytochromes,

leading to a distinct, locally necrosis. Trichloromethyl free radical can also undergo anaerobic

reactions, which may result in the formation of such toxic compounds as chloroform,









CHAPTER 2
MATERIALS AND METHODS

Experimental Animals

Male Fisher-344 rats (Fredricks Laboratories, Fredrick, MD) were used for all experiments

described. The dosage for chronic exposure to either allyl alcohol (AA) or carbon tetrachloride

(CCl4) (Sigma, St. Louis, MO) was determined mathematically calculating the LD10 dose.

Chronic dose of AA consisted of 0.001 mL/kg (7.4 mg/kg) body weight in a dilution of 1:50

vol/vol in 0.9% saline solution. Chronic dose of CCl4 WAS given as 0.38 mL/kg (300 mg/kg) of

body weight in a 1:1 vol/vol dilution in corn oil. These chemicals were administered by i.p.

inj section using a pattern of fiye consecutive days of inj sections followed by two days of rest for a

duration of 90 days. They were given this treatment at the same time of the day for everyday

treated. This exposure pattern would closely mimic an average worker's exposure during a three-

month period. The animals were sacrificed 5-6hr after the Einal inj section. Two hours prior to

sacrifice, animals were inj ected with BrdU (50 mg/kg body weight) for determination of DNA

synthesis as described by Lindroos et al. [50]. All procedures regarding animals were conducted

according to institutionally approved protocols.

Animal Sacrifice and Tissue Collection

All animals utilized for tissue collection were euthanized by administration of an overdose

of pentobarbital (1.0 mL/100g body weight, Sigma). This is consistent with the

recommendations of the panel on euthanasia of the American Veterinary Medical Association

and the Guide for the Use and Care of Laboratory Animals (U. S. Department of Health and

Human Services/NIH Publication #86-23). After euthanasia, sections of liver were collected for

paraffin embedding. These sections were Eixed O/N in 10% Neutral Buffered Formalin (Richard-

Allan Scientific, Kalamazoo, MI). The formalin was then exchanged for PB S and embedded in









of extracellular TGFP from the ECM. To date, five TGFP receptors have been identified

(TGFPR1-5). TGFPR1 is expressed predominantly by hematopoietic progenitor cells, where it

has been shown to form a heterotetramer with TGFPR2 [127].

Activation of Transforming Growth Factor Beta

As mentioned previously, TGFP is synthesized in a latent precursor form that is

subsequently cleaved to the 112 amino acid active form [128]. TGFPl is considered to be the

predominant isoform of the molecule in most cell types and is highly conserved among

mammalian species. Unless specified otherwise, TGFP will refer to TGFP type 1 throughout the

remainder of this report.

N-terminus cleavage of the TGFP pro-peptide occurs within the Golgi apparatus. The

truncated C-termini dimerize via disulphide bond formation between conserved cysteine

residues, giving rise to the secretable, latent form [129]. Further cleavage by extracellular

proteases generates the biologically active form of the molecule. Activated TGFP readily forms a

complex with a2 macroglobulin in serum, and is rapidly cleared by the liver resulting in a tl/2 Of

only 3 minutes [130]. The latent form does not complex with a2 macroglobulin and is able to

persist for a much longer time in serum [131].

TGFP displays a high affinity for the type II receptor [132]. TGFP binding to the type II

receptor mediates the formation of a heterotetramer composed of two type I and type II subunits.

Constitutively active kinase activity associated with the type II receptor trans-phosphorylates

specific serine residues on the type I receptor, initiating an intracellular signaling cascade which

is propagated to the nucleus [132].

The activated TGFP receptor tetramer phosphorylates members of the Smad protein

family, likely Smad 2 and 3. These proteins form a complex with Smad 4 within the nucleus and

recruit additional proteins that form a transcription complex that initiates gene transcription.
















































Figure 3-5 AFP+ hepatic oval cells in the portal regions of rat liver. (Al-3) AA injury at dl5, 45
and 90, respectively. (Bl-3) CCl4 injury at dl5, 45 and 90, respectively. (C) NRL.
(D) immunoglobulin control. (E-F) Higher magnification of d90 AA and D90 CCl4,
respectively. A and B original magnification=600X, C-F original
magnifi cati on= 1000X.









74




Full Text

PAGE 1

COMPARISON OF EXTRACELLULAR MATRIX PROFILES IN ALLYL ALCOHOL AND CARBON TETRACHLORIDE CHRONIC LIVER INJURY MODELS By ALICIA RENAE BROWN A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2008

PAGE 2

2 2008 Alicia Renae Brown

PAGE 3

3 To my parents, Gary and Julie Brown.

PAGE 4

4 ACKNOWLEDGMENTS I express my most sincere gr atitude to my mentor Dr. Bryon Petersen for his great perspectives on life and science. I also thank my committee members for all of the insights on my project. The post doctorates in my lab, (Drs. Seh-Hoon Oh, Li ya Pi, Anna Piscaglia, Thomas Shupe, and Jennifer Williams) have been a tremendous help to my success. My peers (Houda Darwiche, Susan Ellor, Dana Pi ntilie, and Nicole Steiger) ga ve me experimental advice, encouragement, and friendship over the past 2 years. I would also like to thank my parents, Gary and Julie Brown for always supporting everything I have done, and my sisters Christina and Andrea for their consistent support.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........8 LIST OF FIGURES................................................................................................................ .........9 LIST OF ABBREVIATIONS........................................................................................................10 ABSTRACT....................................................................................................................... ............12 CHAPTER 1 LITERATURE REVIEW.......................................................................................................14 Anatomy and Physiology of the Liver....................................................................................14 Cells of the Liver............................................................................................................. .......16 Hepatocytes.................................................................................................................... .16 Endothelial Cells.............................................................................................................17 Immune Cells...................................................................................................................17 Stellate Cells................................................................................................................. ...18 Biliary Cells.................................................................................................................. ...20 Oval Cells..................................................................................................................... ...21 Liver Regeneration............................................................................................................. ....23 Overview of the Extracellular Matrix.....................................................................................27 Alpha-Smooth Muscle Actin..................................................................................................28 Proteins of the Extracellular Matrix.......................................................................................29 Collagens...................................................................................................................... ...29 Collagen Type IV............................................................................................................29 Fibronectin.................................................................................................................... ...30 Laminin........................................................................................................................ ....31 Basement Membranes in the Liver.........................................................................................32 Introduction to Liver Fibrosis.................................................................................................32 Chronic Allyl Alcohol Exposure............................................................................................35 Chronic Carbon Tetrachloride Exposure................................................................................37 Degradation of Pre-Existing Matrix.......................................................................................39 Overview of the Transforming Growth Factor Beta...............................................................40 Activation of Transforming Growth Factor Beta...................................................................41 Biological Activity of Transfor ming Growth Factor Beta.....................................................42 Overview of Connective Tissue Growth Factor.....................................................................43 Biological Significance of Conn ective Tissue Growth Factor...............................................44 Specific Aims of this Study....................................................................................................46 2 MATERIALS AND METHODS...........................................................................................48

PAGE 6

6 Experimental Animals........................................................................................................... .48 Animal Sacrifice and Tissue Collection.................................................................................48 Protein Isolation, Purifica tion and Quantification..................................................................49 Fresh Normal Rat Liver Extraction.................................................................................49 Formalin Fixed Paraffin Embedded Tissue Extraction...................................................49 Acetone Precipitation......................................................................................................51 Protein Quantification with DC Protein Assay...............................................................51 Running Gel.................................................................................................................... .51 Stacking Gel................................................................................................................... .52 Western Blot Analysis.......................................................................................................... ..52 Protein Sample Preparation.............................................................................................52 Electrophoresis of the Western Gel.................................................................................52 Transferring of a Western Gel to a PVDF Membrane....................................................52 Coomassie Staining of Western Gel................................................................................53 Silver Staining of Western Gel........................................................................................53 Probing of Western Membrane.......................................................................................54 Developing of Western Memb rane with ECL Plus.....................................................54 Membrane Stripping for Reprobing................................................................................54 Nucleic Acid Isolation, Puri fication and Quantification........................................................55 RNA Isolation and Analysis............................................................................................55 Reverse Transcription PCR.............................................................................................55 DNA Isolation and Analysis............................................................................................56 PCR Analysis...................................................................................................................56 Histological Analysis.......................................................................................................... ....57 Hematoxylin and Eosin Staining of Paraffin Embedded Tissue.....................................57 Gomori's Trichrome.........................................................................................................58 Chromagen Staining........................................................................................................58 Antigen Retrieval.............................................................................................................59 Immunohistochemistry Analysis............................................................................................59 3 RESULTS........................................................................................................................ .......61 Optimizing Protein and Nucleic Acid Extr actions from Archival Formalin Fixed Paraffin Embedded Tissue..................................................................................................61 Assessing the Allyl Alcohol and Carbon Tetr achloride Chronic Liver Injury Models..........61 Optimizing Protein and DNA Ex tractions from Archival Formalin Fixed Paraffin Embedded Tissue................................................................................................................62 Overall Evaluation of the Two Injury Models........................................................................62 Hepatocyte and Non-Parenchymal Cell Pro liferation in Chronic AA and CCl4 Models......63 Alpha Fetoprotein Expression by Hepatic Oval Cells in Chronic AA and CCl4 Models.......63 Trasnsforming Growth Factor Beta Expr ession by Hepatic Cells in Chronic AA and CCl4 Models........................................................................................................................64 Connective Tissue Growth Factor Expressi on by Hepatic Cells in Chronic AA and CCl4 Models......................................................................................................................... ........64 Alpha-Smooth Muscle Actin Expression by He patic Stellate Cells in Chronic AA and CCl4 Models........................................................................................................................65 Collagen Expression by Hepatic Cells in Chronic AA and CCl4 Models..............................65

PAGE 7

7 Laminin Expression by Hepatic Cells in Chronic AA and CCl4 Models...............................66 Fibronectin Expression by Hepatic Cells in Chronic AA and CCl4 Models..........................67 4 DISCUSSION..................................................................................................................... ....83 Analysis of Archival Tissue....................................................................................................83 Immunohistochemical Analysis of Archival Tissue...............................................................84 5 FUTURE DIRECTIONS........................................................................................................92 LIST OF REFERENCES............................................................................................................. ..93 BIOSKETCH...................................................................................................................... .........107

PAGE 8

8 LIST OF TABLES Table page 2-1 List of antibodies utilized............................................................................................... ....60 3-1 Proliferating hepatocytes and non-hepato cytes over the time course of chronic AA and CCl4 liver injuries........................................................................................................68 3-2 Activation of HSCs over the time course of chronic AA and CCl4...................................68

PAGE 9

9 LIST OF FIGURES Figure page 1-1 Hepatic microarch itecture and blood flow.........................................................................47 1-2 Liver acinus.............................................................................................................. ..........47 1-3 Growth of remaini ng three liver lobes after partial hepatectomy in the rat...................47 1-4 Amount of resident hepatic cells w ithin the cell cycle during the time following partial hepatectomy............................................................................................................47 3-1 Protein and Nucleic Ac id analysis of AA and CCl4 archival paraffin embedded formalin fixed tissue..........................................................................................................69 3-2 Hematoxylin and Eo sin staining of AA and CCl4 treated rat livers 90 days following initiation of injury........................................................................................................... ...70 3-3 Proliferating cells over the time course of chronic AA and CCl4 liver injuries................72 3-4 Immunohistochemical labe ling of proliferating, BrdU+ nuclei..........................................73 3-5 Alpha Fetal Protein+ hepatic oval cells in the portal regions of rat liver...........................74 3-6 Transforming Growth Factor Beta expression during chronic liver injury.......................75 3-7 Connective Tissue Growth Factor expression during chronic liver injury........................76 3-8 Alpha-Smooth Muscle Actin expr ession during chronic liver injury................................77 3-9 Activated HSCs over the time course of chronic AA and CCl4 liver injuries...................78 3-10 Trichrome staining for collage n I during chronic liver injury...........................................79 3-11 Collagen type IV in chronic live r injury models by immunohistochemistry.....................80 3-12 Laminin deposition in chronic liver in jury models by immunohistochemistry.................81 3-13 Fibronectin in chronic liver in jury models by immunohistochemistry..............................82

PAGE 10

10 LIST OF ABBREVIATIONS 2AAF 2-Acetylamino fluoride ADH Alcohol dehydrogenase ALDH Aldehyde dehydrogenase AA Allyl alcohol AFP Alpha fetal protein -SMA Alpha-smooth muscle actin APC Antigen presenting cell BM Basement membrane CCl4 Carbon tetrachloride ColI Collagen type I ColIV Collagen type IV CTGF Connective tissue growth factor CYP2El Cytochrome P4502E1 EC Endothelial cell EGF Endothelial cell growth factor EMT Epithelial-mesenchymal transition ECM Extracellular matrix FN Fibronectin GFAP Glial fibrillary acidic protein GSH Glutathione HOC Hepatic oval cell HSC Hepatic stellate cell HGF Hepatocyte growth factor IGF Insulin-like growth factor

PAGE 11

11 KC Kupffer cell LM Laminin LRP Low density lipoprotein receptor MMP Matrix metalloproteinase MAPK Mitogen-activa ted protein kinase NASH Nonalcoholic steatohepatitis NFB Nuclear factor kappa B PH Partial hepatectomy PDGF Platelet-derived growth factor beta PI3K Phosphatidylinositol 3-kinase SEC Sinusoidal endothelial cell TIMP Tissue inhibitor of metalloproteinases TGF Transforming growth factor beta TGF R Transforming growth factor beta receptor TNF Tumor necrosis factor alpha vWF von Willebrand factor

PAGE 12

12 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science COMPARISON OF EXTRA CELLULAR MATR IX PROTEINS FROM ALLYL ALCOHOL AND CARBON TETRACHLORIDE By Alicia Renae Brown May 2008 Chair: Bryon Petersen Major: Medical Sciences Allyl alcohol (AA) and carbon tetrachloride (CCl4) are two chronic liver injury models that cause extensive periportal and centrilobular damage, respectfully. The extracellular matrix (ECM) is a complex structure aiding in cell ac tivation, migration and di fferentiation within the liver. The composition of this ma trix varies throughout the liver corresponding to cellular and functional requirements within ea ch region of the organ. During th e liver regeneration process, the ECM goes through substantial changes which can provide evidence to the signals required for restoration of the liver mass. Transforming gr owth factor-beta and co nnective tissue growth factor have been shown to be ke y cytokines in the regul ation of this process. Cells attach to the ECM by means of transmembrane glycoproteins cal led integrins. The extracellular portion of integrins binds to various types of ECM proteins including collagens, laminins and fibronectin. The current study examines the ro le of molecular signals and EC M components in each of these two chronic injury models. Archival, formalin fixed paraffin embedded liv er tissue from AA and CCl4 treated rats was examined. Attempts at nuc leic acid and protein extraction from these tissues were unsuccessful. Theref ore, immunohistochemistry was us ed to describe the fibrotic response. AA and CCl4 proved to be very different models of chronic liver disease, as one resulted in massive necrosis (AA) while the other resulted in cirrhosis (CCl4) as evidenced of

PAGE 13

13 Trichrome staining. The AA model deve loped a lose ECM at the site of injury that was devoid of collagen type IV, where as CCl4 developed a dense fibrotic scar that was rich in collagen type IV. Understanding the composition of ECM during chronic liver in jury could lead to better methods for the treatment of pathol ogies involving hepatic fibrosis.

PAGE 14

14 CHAPTER 1 LITERATURE REVIEW Anatomy and Physiology of the Liver The liver is one of the most intriguing orga ns of the body. It is the largest parenchymal organ; an adult human liver can hold about 1.5L of blood which is close to 25% of cardiac output per minute [1]. About 30% comes from oxygen ric h, nutrient poor arteries; the rest comes from the nutrient rich, oxygen poor portal vein. The liver's efficient, dual afferent blood supplies to maintain its high metabolic activities. Blood from both the hepatic artery and the portal vein enter proximal to each other next to bile ducts in a place termed the portal triad. Blood flows in from here and is filtered through plates of hepato cytes and drains into th e central vein. Sinusoids composed of endothelial cells line each cord of hepatocytes and enclose the micro-vascular circulatory system of the liver. Blood flows into the hepatic artery and portal vein, filtering through the hepatic plates to th e central vein. Bile flows in the opposite dire ction receiving components from the hepatic plates, and then drai n through the bile ducts towards the gal bladder where it is stored (Figure 1-1). Bi le is drained into the digestive tract to act as a detergent [2]. Three interpretations of this design exist include the cla ssic lobule based on structural parameters, a portal lobule for bile drainage, and an acinus concept fo r oxygen gradients. The acinus concept is the most common because it pr ovides the greatest functional explanation [3]. Cells nearest the portal triad are in zone 1 which has access to the highest concentration of nutrients and oxygen, while cells near est the central vein are in z one 3 with the least amount of nutrients and oxygen (Figure 1-2). The unidirecti onal perfusion of blood in sinusoids creates different microenvironments for hepatocytes near the periportal venular in lets versus terminal hepatic venular outlets [4]. Interestingly, hepa tocytes within Zone 3 have an increased DNA content (4N to 16N); predominant bi-nucleation and large in size si milar features are seen in the

PAGE 15

15 Zone 2 cells which are typically 16N. Conversely, hepatocytes w ithin Zone 1 are smaller and usually single nucleated (2N) [5]. The many functions of the liver involve main taining homeostasis of the body, metabolizing amino acids (forming ammonia and converting it to urea), lipids, xenobiotics, serum proteins and carbohydrates, converting glucose into glycogen for storage. The liver is also one of the main sites for insulin and glucagon degradation which assists the pancreas control blood glucose levels. In animals, regulation of blood glucose levels by the liver is a vital pa rt of homeostasis. In hepatocytes, any extra glucose is stored as glycogen through phosphorylation into glucose-6phosphate (G6P) and, when needed, glucose can be released into the bloodstream through G6Pase. In addition, the liver maintain s the colloid osmoti c pressure of the blood by producing the most abundant plasma protein, al bumin. It also produces other v ital plasma proteins such as lipoproteins, glycoproteins including prothrombin and fibrinogen, and the nonimmune and globulins. Although the liver does not actively produce hormones, it modifies the actions of hormones released by other organs. The liver stor es and converts vitamins A, D and K taken up from the blood stream to a more functional form. St ellate cells store vitamin A as retinyl esters within their lipid pools. Vitamin A is important for vision as it is a component of rhodopsin (a pigment in the rods and cones of the eye) a nd for proper bone growth among others. Vitamin D3 (cholecalciferol) is generated in the skin of animals when light energy is absorbed by a precursor molecule 7-dehydrocholesterol. Cholecalcifera l is hydroxylated to 25-hydroxycholecalciferol by the enzyme 25-hydroxylase (CYP27, which is found in high concentrations in the liver) then it goes to the kidney to be further modified in to the active form of 1,25-dihydroxycholecalciferol

PAGE 16

16 which is involved in mineral metabolism and bone growth. The liver utilizes vitamin K for the production of clotting factors and synthesizing vitamin K-dependent coagulation factors. Processing large volumes of blood induces the liver to function as a detoxifying organ. Liver enzymes, such as alcohol dehydrogenase (ADH), cytochrome-P450 (CYP) and isoforms of uridine diphosphoglucuronate gl ucuronosyltransferase (UGT) a llow for the alteration of chemical composition of many xenobiotics and th eir subsequent removal. The conversion of nonhydrophilic drugs to a more water soluble form ai ds in their excretion by the kidneys occurs in the liver. Homeostasis of blood ir on levels depends directly on the ability of the liver to store and metabolize iron. To perform all of these different functions, the liver requires many different cell types which include hepatocytes, bile ductular epithelia l cells, stellate cells, si nusoidal and vascular endothelial cells, liver specific m acrophages and immune cells [1]. Cells of the Liver Hepatocytes Hepatocytes have a life span of 300-400 days and are capable of 86 doublings [6; 7]. Their division results in compensatory hyperplasia, which restores the liver mass after injury. The hepatocytes are the functional endo crine and exocrine cells of th e hepatic lobule and make up to 80% of cellular mass of the liver. One of the mo st important functions of hepatocytes is the production of bile. Gapand tight -junctions on the sides of hepatocyte membranes help to distinguish the basolateral from the apical membrane domain. The basolateral domain has abundant microvilli which project into the space of Disse to maximize surface area to allow optimum exchange between hepatocytes and bl ood [1]. Hepatocytes communicate through these connections with the sinusoids, allowing cross-ta lk with each other and additional cell types. Cyclic nucleotides, inositol 1,4,5-trisphosphate, a nd calcium can enter through gap junctions to

PAGE 17

17 lead to the propagation of calci um waves along the intact hepatic acinus, enabling a coordinated hepatocellular response to exogenous stimuli [4]. With fibrosis, hepatocytes lose some of their microvilli and the fenestrae of endothelial cells close, dwindl ing communication [Friedman, 1993]. Endothelial Cells Sinusoidal endothelial cells (SEC) are the ma in cell that makes up the sinusoid lining of the liver. They add up to about 20% of the total number of cells in the liver. Being one of the first cells exposed to xenobiot ics in the blood they are primar y response cells, changing shape and sending signals. SECs have been characteri zed as a unique type of endothelial lining consisting of endothelial cells with flattened pr ocesses perforated by small fenestrae of about 100nm. It is well known that fenestrae diameter and porosity decreases from the pericentral to periportal regions; zone 3 has the highest fluid exchange capability [8]. This pore size is regulated by many endogenous factors and xenobio tics. SECs also posse ss an extensively invaginated plasma membrane, numerous vesicles and lysosome-like vacuoles, indicating a high endocytotic activity [8]. Immune Cells Kupffer cells (KC) were first observed by Karl Wilhelm von Kupffer in 1876. KCs make up the largest population of tissue macrophages [9]. These cells dwell within the hepatic sinusoids as the resident macropha ges of the liver. KCs contain ma ny invaginations with proteins and carbohydrate molecules along the cell membrane putatively to enhan ce its functions. KCs clear viruses, old red blood cells, bacteria a nd other foreign materials through endocytosis. KCs send out signals when they are activated (typica lly by neutrophils and reactive oxygen species) to prepare the liver to isolate these substances producing free radicals, oxidants (via NADPH oxidase) and cytokines such as tu mor necrosis factor alpha (TNF ). Oxidants activate KC NF-

PAGE 18

18 B, causing an increase in TNF production. TNF induces neutrophil infiltration and stimulates mitochondrial oxidant production in hepatocytes, which are sensiti zed to undergo apoptosis [10]. KCs also release ecosinoids and leukotrine B4 which are known chemotactic factors for neutrophils [11]. KCs highly express F4/80 a nd Major Histo-Compatibility class I (MHC-I). Expression of MHC-II, CD1d, and CD86 are det ected only at low amounts [12]. KCs produce IL1, IL6, TGFand latent TGF binding protein 1, 2 [13], TGF R [14], matrix metalloproteinases 2, 9, 13, 14 and th eir inhibitors, TIMP-1 [15] a nd TIMP-2 [16]. This triggers stellate cells to lay down a protective layer of proteins, consisting of mainly collagen. Pitt cells are attached to endothe lial lining of the sinusoids. Pitt cells are granular cells with a high natural killer cell activity, putatively to act ag ainst tumor and virus-infected cells [1]. Mast cells are a rich source of nerve growth factor a nd other cytokines to dest roy foreign materials and accumulate in fibrotic liver. Other blood cells in the liver include neutrophils, T, natural killer (NK) and NKT cells. Neutrophils ar e present in the sinusoids of th e liver and migrate to the site of inflammation and act as shor t-lived macrophages (1-2 days). Neutrophils and T cells express receptors for TGF and INF attracting cytokines (through ch emotaxis) that are produced by KCs, mast cells, ECs and by the ne utrophils [17; 18]. Neutrophils ar e the first responders in an injury; they also can activate KCs which then notif y the neutrophils and othe r cells (i.e. stellate cells) to release more factors. Stellate Cells Star-like cells were also first observed by Karl Wilhelm von Kupffer in 1876 and were shown to reside in the space of Disse and have a nucleus of about 40um. They were termed hepatic stellate cells (HSC) and represent 5-8% of normal liver cells [1]. These cells store vitamin A (nearly 90% of the total amount in the body) and lipids. Since vitamin A is lipid soluble this yields a very complimentary situat ion. They produce little amounts of collagen in

PAGE 19

19 normal liver [19]. They are also defined by expression of ectodermal neural markers such as glial fibrillary acidic protein (GFAP) [20]. HSC numbers increase significantly upon liver damage [2 1]. Liver fibrosis is a common consequence of chronic liver diseases and is an indirect result of the ac tivation of HSCs. After liver damage, HSCs undergo a transition from a quiescent to an activated phenotype. The storage of fat or vitamin A dwindles, leading to vita min A deficiencies and associated disorders. Activated HSC express -smooth muscle actin ( -SMA), and multiple chemokines and growth factor receptors, including, lept in, endothelin and platelet-deriv ed growth factor beta (PDGF) [21]. Activated HSCs upregul ate the expression of TGF and establish an autocrine-loop whereby this cytokine upregulat es the expression of collagen type I genes by transcriptional mechanisms involving peroxides [22; 23]. Both smooth muscle associat ed proteins are only expressed on HSCs when they are activated th rough liver injury. This is for contractile capabilities and regulati ng portal blood flow. Although HSCs ma y be the most important source of myofibroblasts in injury, HS Cs are not the sole source. The portal area of the normal liver contains fibroblast-like cells (myofibroblasts), an d the proximal bile ductu les, up to the canals of Hering, are encircled by smooth mu scle cells, which may proliferat e, migrate and contribute to ECM production in respons e to biliary injury. In addition to actin, activated HSCs produce and deposit structural proteins such as collagen types I, III and IV, tenascin, indulin, la minin, entactin and perlecan directly into the space of Disse [21]. These proteins are deposited in areas of inflammation in an effort to contain the spread of infection. Until the 1980s, hepato cytes where considered the main source of extracellular matrix, but in fact HSCs are the main culprits. However, they can also produce proteinases, MMPs that degrade the ECM all while also producing their i nhibitors TIMP-1 and -

PAGE 20

20 2. The main role in fibrosis reduction by HS Cs is by their apoptosis and reduced hepatic expression of metalloproteinase inhibitors which results in sp ontaneous resolution [24]. HSCs also produce growth factors such as TGF connective tissue growth f actor (CTGF), hepatocyte growth factor (HGF) and epithelia l growth factor (EGF) [21]. TGF is negated by TNF and if it is blocked, not only are stellate cells are not activated, but the liv er is protected from ethanolinduced hepatic injury [25; 13]. HSCs may function as professi onal APCs for activation or re stimulation of T cells through the space of Disse when the endothelial cell barr ier is injured. HSCs activate T cell responses and express CD31 on their surface at a high de nsity and present anti gens to MHCIand IIrestricted T cells for cleara nce [12]. According to Winau et al HSCs (with > 98.5% purity) are capable of inducing vigorous NKT cell responses in vitro and in vivo and promoting homeostatic proliferation of NKT cells th rough production of IL-15 [12]. Biliary Cells Cells of bile ducts are commonly called biliary cells, bile duct cells or cholangiocytes. Bile is formed and secreted by hepatocytes and flows opposite of the blood flow through the bile caniculi into the canals of Herring and then to the bile ducts flowing into the larger common hepatic duct. If the animal has a gal bladder it would then be stored here until needed, otherwise it is immediately released down the common bile duct to the duodenum. If this flow is disturbed, bile can build up into the liver and cause infla mmation, fibrosis and eventually cirrhosis. Biliary cells produce and express collagen type IV, la minin, entactin and perl ecan. These cells also produce the growth factors TGF [26] and CTGF [27]. Oval cells reside in the same space, most likely further down in the canal of Herring and are drawn into th e liver through injury, and can form more biliary cells or other liver cells in vivo Biliary cells can also arise from hepatocytes,

PAGE 21

21 proved by Michalopoulos in 2005 when he reported the findings from transplanted hepatocytes into a bile duct ligated rat [28]. Oval Cells In 1956 E. Farber recognized the same cell type appearing in the liver after several different chemical injury models. He termed th ese small cells with high nuclear to cytoplasmic ratios as oval cells. Although debate continues as to the site of origin of these cells, they are universally considered to be the resident hepatic stem cell [29]. Stem cells are defined as cells that are undifferentiated, capable of self-renewal, with potential to differentiation into multiple lineages and having the flexibility to use all of these options. Oval cells are recognized as playing an important role in th e etiology of hepatic gr owth and development [30]. They have the ability to proliferate clonogenica lly and differentiate into severa l lineages, including bile ductal epithelia, hepatocytes, intestinal ep ithelia, and exocrine pancreas [31]. In normal liver tissue, oval cells are almost beyond detection; however, stem cell activation leads to the profuse replication of these cells in the periportal re gions of the liver. Oval cells are more readily seen after a severe liver injury in conjunction with hepatocyte inhibition such as the 2AAF model [32]. This models shows that these cells expand fr om the Canal of Hering [29], from a stem cell niche and/or from the bone ma rrow [33], and enter hepatic parenchyma in the periportal regions where oval cells proliferate and differentiate into hepatocytes and bile duct cells [34]. When the bile ductular epithelia are damaged in periportal zones, oval cell proliferation is reduced [35]. Morphologically, oval cells are sm all in size (approximately 10 m), with a large nuclear to cytoplasmic ratio and an ovoid nucleus, ther eby giving them their name. Oval cells are a heterogeneous cell population with several phenotypes. Oval cells possess characteristics similar

PAGE 22

22 to ductular cells in their distinct isoenzyme profiles intermediate filaments (BSD7, OC2, OC3, OV-1, and OV-6), extracellular matrix proteins (CK8, 18, 19), enzymes a nd secreted proteins (alpha-fetoprotein, -glutamyl transferase) [36; 37; Na gy 1998]. They also possess hepatocyte features such as producing albumin. They express some hematopoietic stem cell markers such as CD-34, c-kit and Thy1 as well as Flt-3 [39; 40; 41 ; 33]. When they are pres ent in the liver they differentiate toward both hepatic and bile ductular epithelial lineages. When going towards hepatocyte formation, it was shown that oval ce lls first differentiate into basophilic small hepatocytes and then into matu re adult hepato cytes [42; 43]. Oval cells may constitute more than 50 % of the liver during regeneration caused by administration of 2AAF/PH; it is thought that the cells form a transit amplifying compartment that includes undifferentiated pr ogenitors, medially-different iated transit cells, and newly differentiated hepatocytes [44]. Oval cells are sim ilar to hepatocytes in that both require growth factors for cell cycle progression, an d both cell types also require a priming process in order to respond to these stimuli [45]. Oval cells expre ss c-met, the receptor for HGF and therefore are growth responsive during the time of regenerati on when HGF levels are high [46]. Instead of oval cells simply creating numerous bile ducts, there is morphologi c evidence that these cells are potentially going through a preneoplastic process. Petersen et. al exposed rats to methylene dianaline (DAPM) 24hrs prior to hepatic damage (2-AAF/hepatic injury, PH or CCl4) [35]. Under these circumstances the bile ductular epithelium was destroyed and the oval cell response was seve rely inhibited. This study was the first to elucidate a direct association between the requirem ent of an intact bile ductular epithelium and the ability to mount an oval cell response. This da ta, however, does not prove that oval cells arise from bile ductular cells because the DAPM could have elicited either a dire ct or indirect toxic

PAGE 23

23 effect upon the oval cells, which may have resulted in the inhibition of th eir activation. This study does support the idea that oval cells arise from the periportal zone, specifically from the canals of Hering [35]. Petersen re ported in 1999 that oval cells also come directly from the bone marrow [47]. Liver Regeneration Under normal conditions, hepatocytes exhibit no minal replicative activity; only 1 in every 20,000 hepatocytes undergoes mitotic division at any one given time point. Yet, hepatocyte division is the major driving fo rce behind liver regeneration [48] Compensatory hyperplasia of the liver (Figure 1-4), takes plac e in response to mild to severe liver injury resulting from surgical resection of a portion of the liver or exposure to destruc tive agents such as hepato-toxins or hepatotropic viruses. This te rm is utilized instead of rege neration due to the unusual response to gain enough tissue to function normally. Instea d of gaining individual lobules back it will form one large lobule. The liver must be a certai n size to do all the jobs necessary to keep at homeostasis. In 1931 Higgins and Anderson repo rted the first partial hepatectomy (PH), removing 69% of the liver in a rat [49]. Removal of as much as 80-90% of the liver can be restored in the absence of disease [1]. Rats have been subjected to 2/3 hepatectomy (12 times in all) with the liver able to rege nerate itself. Each time, the liver is restored to its normal size within a few weeks (Figure 1-3). It has been estimated that one rat hepatocyte has the capacity to generate at leas t 50 livers [30]. In the rat, hepatocytes move from the G0 resting phase of the cell cycle into G1, as mediated by the cyclin D1 pathway within 15hrs of PH. Periportal hepato cytes are the first to undergo DNA synthesis and prolifer ation gradually spreads to in clude the hepato cytes located around the central vein. A large peak of DNA synthesis is observed at about 24hrs post PH, and a second, yet smaller peak at 48hrs. The smaller peak reflects DNA synthesis occurring in non-

PAGE 24

24 paranchymal cells (NPC) and pericentral hepato cytes (Figure 1-4). Unlike hepatocytes that display a wave of DNA synthesis from periportal to pericentral, NPCs across the lobule exhibit simultaneous DNA synthesis. The original liver ma ss is usually restored within 10 days of the hepatectomy. Liver regeneration involves a tightly regulation of cells to enter cell cy cles after injury and to exit cell cycles after appropr iate tissue remodeling. In the hepa tocyte-driven injury response, almost all types of cells in liver proliferate at least once with some cells dividing two times after the completion of regeneration. He patocytes are the firs t to enter DNA synthe sis in 12hrs after 2/3 PH injury detected with bromodeoxyuridine labeling, followed by HSCs, KCs and then ECs as seen in figure 1-4. With organ maturity, the pr ocess of differentiation leads to the commitment of differentiated cells to constitu tive functions that maintain ho meostasis and specialize functions that serve organisms needs. In the mature livers of all species, prolifer ation of all cell types subsides to a low level, thus, the mature liver co nsists of two types of cells: intermediate cells, the hepatocytes, which replicate infrequently, but can respond to signals for replication, and replicating cells, the stem cells, SECs, KCs, and HSCs, bile duct epithelium, and granular lymphocytes (pit cells). The earliest event, occurring w ithin one minute after, is a la rge increase in the blood level of hepatocyte growth factor (HGF) released from the remodeling ECM [50]. Active HGF then binds to its receptor c-met to lead the hepato cytes into the cell cy cle. The amount of HGF putatively is in an inverse rela tionship with total liver mass of hepatocytes. The earliest event, occurring within one minute after injury, is a large increase in the blood level of HGF released from the remodeling ECM [50]. Several ne gative regulatory signals may include TGF [51],

PAGE 25

25 p53, p21 and C/EBP [52]. Mice lacking p53, p21 and C/EBP show continuous hepatocyte turnover and hyperproliferation of hepatocytes [52]. Other factors such as interleukin 6 (IL -6), tumor necrosis factor-beta (TNF ), epidermal growth factor (EGF) and, an EGF homol og, transforming growth factor-alpha (TGF ) are also involved in this response. TGF mRNA and protein levels increas e markedly within hours after PH [53]. IL-6 and TNF knockout mice both show significantly delayed regeneration after PH [54; 55; 56]. Less knowledge is known about the m echanisms to terminate liver regeneration. Several animal models of chemical hepato toxicity have been developed to study the mechanisms regulating the prolif erative response to liver injury. Among the most extensively utilized chemical agents are carbon tetrachloride (CCl4), which causes necrosis of the centrilobular regions of the liver, and allyl alc ohol (AA) which causes periportal necrosis. In both models of acute hepatic in jury, regeneration of the necrotic region is mediated by proliferation of mature hepato cytes and the oval ce ll response not activat ed to a degree of importance, if at all. Normally, hepatocyte-driven injury response is e fficient to replace lost cells and serves as a primary source for liver repairs [30]. However, when damage to a liver is too profound or proliferation of hepatocytes is inhibited as a result of, e ither the metabolism of foreign compounds to hepatotoxic intermedia tes, or hepatotropic virus inf ection, progenitor cells such as oval cells are recruited to proliferate and repl enish the function of da maged hepatocytes [36]. Morphologically, oval cells are sm all in size (approximately 10 m), with a large nucleus to cytoplasm ratio, with an oval shaped nucleus [ 29]. After activation, a large number of oval cells appear near bile ductules and th en migrate into the hepatic pare nchyma [57]. Activated oval cells firstly differentiate into bas ophilic small hepatocytes and eventually become mature adult

PAGE 26

26 hepatocytes [29; 42]. Besides hepa tocyte lineage, oval cells are also able to differentiate into intestinal type epithelium in rats in vivo [42], be induced to bile duc tular epithelial cells [58] and pancreatic-like cells in culture [34]. The origins of oval cells remain cont roversial and several key questions regarding the molecular cues that initiate oval cell prolifera tion and direct lineage specific differentiation still remain. In progenitor-dependent injury response, alt hough it is not clear whether the same positive and negative regulatory signals contribute to liver regenerati on, injury-induced changes in cytokines and growth factors defi nitively modulate the fate of the facultative stem cells (oval cells) [59]. A distinct response a ssociated with activa tion of oval cells is a large expansion of desmin-positive stellate cells and an increased production of growth factors and ECM molecules in the periportal regions of injured livers. It is believed that stellate ce lls secrete these growth factors such as HGF, TGF TNF and ECM components such as fibronectin and regulates expansion and differentiation of the oval cell po pulation. In addition, the autocrine production of TGF acidic fibroblast growth factor and insulin-like growth fact or (IGF) II in oval cells is suggested to mediate the progenito r-dependent liver regeneration. Recently, analysis of morphological cha nges using immunoelectromicroscopy and immunostaining reveals that activ ated oval cells form elongated ductular structures, which are surrounded by basement membranes and terminated at hepatocytes located at the limiting plate and accompanied by activated and proliferating st ellate cells. The surrounded base membranes may provide a substrate and scaffold for pro liferation and migrati on of oval cells. More interestingly, a number of pro liferating HSCs are always asso ciated with these ductules and sometimes form direct cell-cell contact with the ductular epithelial cells. This connection

PAGE 27

27 between oval cells and stellate cells may form the structural basis for the cross talk between these two cell types [60]. Overview of the Extracellular Matrix Most of the cells in multicellular organi sms are surrounded by a complex mixture of nonliving material that makes up the extracellula r matrix (ECM). Hepatic ECM that separates parenchyma from sinusoids in the space of Disse is characterized by a ba sement membrane-like matrix with a low density and is critical for ma intaining the differentiated functions of resident hepatic cells. The structure of the ECM in nor mal and fibrotic liver is based mainly upon collagens, proteoglycans and glyc oproteins. This matrix is alte red rapidly to the slightest metabolic changes of the cells producing it and then it in turn alters the biochemical and morphologic phenotype of thos e cells [61]. It can act as a posi tive as well as a negative regulator of functional differentiation depending on the cell type and the genes studied [62]. The ECM, a modulator of many cellular functi ons, controls and main tains the internal environment of all functions and structures wh ile dispensing energy in th e immediate distant area. The matrix binds growth factors and other signaling proteins to itself to create microenvironments [61; 19]. Cells attach to the ECM by means of transmembrane glycoproteins called integrins. The extracellula r portion of integrins binds to various types of ECM proteins, the majority of which is made up from collage ns, laminins and fibronectin. The intracellular portion binds to the actin filaments of the cytoskeleton. Th e ECM components may function as an affinity matrix for binding and immobilizing so luble growth and differentiation factors such as TGF which binds to collagen IV, and control lo cal regulation of thes e functions [63]. HGF and EGF are also major factors in this matrix. HGF is involved in the induction of cell proliferation and mo tility, induction of morphogenesis, stimulation of T cell adhesion to endothelium and migration, enhancement of

PAGE 28

28 neuron survival, and regulation of erythroid differentiation [64] HGF is found in large amounts in the ECM where it can create a local environment and be copious ly released after injury. HGF may inhibit TGF expression [19]. EGF leads to a spread of HSCs within the space of Disse, perpetuating fibrotic disease [65]. A combination of TGF and EGF induce HSCs to migrate [65]. The space of Disse has a low-density ECM that allows rapid bidirectional macromolecular exchange with plasma. Around vessels, the ECM beco mes denser, as it is packed with fibrillar collagens (except around bile ducts), struct ural collagens, laminins, heparin sulfate proteoglycans, fibronectins and integrins. Hepa tic ECM separating parenchyma from sinusoids in the space of Disse is characterized by a base ment membrane-like matrix with a low density and is critical for maintaining the differen tiated functions of resident hepatic cells. Alpha-Smooth Muscle Actin Actin makes up the most abundant protein in ma ny eukaryotic cell types. It polymerizes forming microfilaments that have an array of functions including re gulating contractility, motility, cytokinesis, phagocytosis, adhesion, ce ll morphology, and providing structural support. It binds the intracellular portions of integrins to aid in cell attachment to the ECM. AlphaSmooth muscle actin ( -SMA), an isoform typical of smooth muscle cells (SMC) and present in high amounts in vascular SMC, was demonstrated in the cytoplasm of pe ricytes of various rat and human. In SMC and pericytes, -SMA is localized in microf ilament bundles for contractile functioning [66]. The first 125bp of the -SMA promoter contains three regions that allow TGF -induced activation. A mu tation in any one of these el ements completely abolishes transcriptional activity [67]. HSCs (pericytes) produce -SMA which is deposited in the regenerating and damaged areas. This contractil e protein supports cell motility and acts as a structural support for pr oliferating cells.

PAGE 29

29 Proteins of the Extracellular Matrix Collagens Collagen is the major protein in mammals, maki ng up to 25% of all pr oteins in the body. It exists in many forms to support its unending set of functions. The uniting structural point of all collagens is that almost every th ird residue is glycine. The major functions of collagen are tensile strength, flexibility, motility, adhesion and overall tissue support [19]. The slightest of mutations in any of the collagens result in major defects such as osteogenesis imperfecti. The chief cell type that produces collagen in the liver is th e HSC [21]. Other liver ce lls have been shown to produce small amounts of it as well, including he patocytes and ECs [19]. Collagen accumulation protects the rest of the organ fr om spreading viruses and disease, and can provide a scaffold for regenerating liver cells. However, this accumulation also can cause the cells within to die due to lack of nutrients. Accumulation in the hepatic si nusoids obstructs blood flow which can also lead to portal hypertension, edema and ascites. Scar ring leads to blocked blood flow, decreasing function/processing of nutrients, hormones, dr ugs, toxins and viruses [21]. Since 1986, TGF has been known to be an intrinsic re gulator of collagens [68]. Collagens are dr amatically increased upon chronic injury from chemicals such as CCl4 and AA [19]. The most abundant protein in the mammalian body is the fibrillar collagen type I (ColI). ColI transcript s contain CCAAT binding sequences that are highly conserved between sp ecies [69]. CAAT binding proteins and other transcription factors present in the live r bind these sequences. According to Yang et. al, ColI can serve as a chemoattractant stimulus for HSCs [65]. Collagen Type IV Collagen type IV (ColIV) is non fibrillar collagen composed of at least six different chains which form a sheet (mesh-like) structur e. The most common form of this collagen molecule is the ubiquitous heterotrimer. Its mol ecules interact to form a complex network via

PAGE 30

30 interactions of adjacent molecule s, both laterally and at either terminus via disulfide bonding. Free deposits of ColIV are found in the subsinusoi dal space to increase tensile strength to make up for the lack of basement membrane ar ound the space of Disse [2003, Yang]. It is found throughout the liver, in both the central and periportal zones. Within the ECM, TGF can bind ColIV [48]. Fibronectin In normal liver Fibronectin (FN) is found in th e subcapsular connective tissue, septa, portal areas surrounding interstitial cells and the space of Disse. FN is a glycoprotein that has functions associated with cell adhesion and migration. FN organizes cell-cell intera ctions and cell-ECM contacts by binding to different components of the ECM and to membrane-bound FN receptors (integrins) on cell surfaces [70] It is found in BMs and throughout the ECM. It exists in two forms, termed cellular FN and plasma FN [48]. Cellular FN is made by fi broblasts, chondrocytes, endothelial cells, macrophages, as well as some epithelial cells. In the liver, HSCs and KCs are responsible for its production and deposition. FN is deposited in the ECM as filaments which are significantly insoluble. Plasma FN is synthesized by hepatocyte s and represents about 1% of serum protein which is about 300 micrograms pe r mL [48]. The composition of FN depends on the tissue type and/or cellular conditions. FN is made up of about 5% carbohydrates wh ich bind to proteins such as integrins, collagens, fibrin, heparin and it also binds to CT GF. FN anchors cells to collagens for adhesion [48]. It also alters cell mor phology and surface architecture. The main function of FN is its involvement in cellular migration during develo pment and regeneration, and regulation of cell growth and differentiation [70]. According to Yang et. al, FN can serve as a chemoattractant stimulus for HSCs [65]. FN also is a part of the provisional matrix during oval cell mediated

PAGE 31

31 liver regeneration [70]. FN has been shown to bi nd IGFBP3 and to form ternary complexes with IGF-1 in human plasma to possibly se quester them into tissues [71]. Laminin Laminin (LM) is a large ECM glycoprotein found in basement membranes of epithelia, surrounding blood vessels and nerves in establishe d tissues. This protein interacts with cell surface receptors and has roles in cell migration during embryonic development and tissue organization [72]. LM subunits (A, B1, B2, S and M) in the perisinusoidal space of the rat liver have been detected in small streaks of baseme nt membranes extending from the portobiliary tract and to a lesser degree from the central vein and perisinusoidal spaces [72]. LM is co-expressed with ColIV. LM is also a prominent component of the perisinusoidal matrix during development, injury and regeneration of the liv er. It is present within the cy toplasm of hepatocytes, SEC and HSC during these times [19]. In fibrosis, the HSC are the main contributors of laminin. Due to its many isoforms of each subunit, it is difficult to view every possible place it is located. LMs are secreted and incorporated into cel l-associated extracellular matrices. These proteins form independent networks associated with ColIV via nidogen [73] and through light, direct interactions with each other [74]. LMs also bind to cell membranes through integrin receptors and other plasma membrane molecules, such as the dystroglycan glycoprotein complex and Lutheran blood group glycoprotein [75]. Th rough these interactions, laminins critically contribute to cell attachment a nd differentiation, cell shape and m ovement, maintenance of tissue phenotype, and promotion of tissue survival. It has been shown in vitro that hepatocytes have an increased affinity for laminin post PH [76]. The main role of laminin may be to provide cell attachment for proper tissue or ganization during liver regenera tion [77; 75]. Work done by Shakado in 1995 showed that laminin was crucial fo r SECs to form long tubular structures while advancing into a gel matrix [75].

PAGE 32

32 Basement Membranes in the Liver Basement membranes (BM) are considered to be a part of the ECM. Nerves, bile ducts, arterial, and venous vessels have BM. In c ontrast, sinusoids are devoid of BM. Although perlecan and some Col IV are f ound in the space of Disse, neither la minin nor entactin is present, suggesting that at least one of these components is needed for the assembly of a BM. In other glandular structures, such as th e thyroid and pancreas endothelial and the epithelial BMs exit. The hepatocyte is the only epithelial cell in the body not separated from the vascular space by two continuous BMs [1]. BMs allow a rapid bi directional macromolecular exchange between plasma and hepatocytes, but are no t as tolerant as the rest of the ECM. The formation of BM-like structures in the space of Disse is charact eristic of cirrhosis, a marked decrease in macromolecular exchange which leads to cell st arvation, toxin buildup, and ultimately cell death [48]. BM matrix integrity, compos ition, and cell-matrix interactions play an important role in anchoring HSCs and preventing them from spread ing within the space of Disse and potentially elsewhere in the liver [65]. Introduction to Liver Fibrosis Hepatic fibrosis is a scarring process that is associated w ith an increased and altered deposition of extracellular matrix in the liver [78]. It was historically thou ght to be a passive and irreversible process due to the collapse of th e hepatic parenchyma leading to septa-forming condensation of pre-existing stroma However, liver fibrosis is an active wound-healing process. Generally, renewing cells are more vulnerable to chemical injury than intermediate cells, which are largely quiescent. The main causes of liver fibrosis include chronic HCV infection, alcohol abuse, and nonalcoholic steatohepatitis (NASH). Chronic hepati c inflammation is tightly linked to fibrosis in virtually all individuals with liver disease and in experiment al models of fibrogenesis [79].

PAGE 33

33 Fibrosis is a well-known histologic al and biochemical hallmark of ci rrhosis, but fibrosis does not necessarily accompany cirrhosis. Originally fibrosis was defined by a WHO expert group in 1978 as the presence of excess collagen due to new fiber formation [80]. Early clinical reports in the 1970s suggested that advan ced liver fibrosis is potentially reversible. Even with this known, liver fibrosis received lit tle attention until the 1980s, when HSCs were identified as the main collagen-producing cells in the liver [81]. The accumulation of ECM proteins alters the hepatic architecture by forming scar tissue. The subsequent development of nodules of re generating hepatocytes defines cirrhosis. The nodules of a cirrhotic liver lack normal lobular organization and are surrounded by fibrous tissue which is formed by a combination of necrotic cellular debris and de novo ECM protein. Cirrhosis produces hepatocellular dysfuncti on and increased intrahepatic resistance to blood flow, which result in hepatic insufficiency and portal hypertension, respectively. In cirrhosis, increased pressure in the portal vein causes large vein s (varices) to develop across the esophagus and stomach to bypass the blockage. The varices beco me fragile and can bleed easily. The essential features of cirrhosis are considered to be parenchymal necrosis, regeneration, and diffuse fibrosis, resulting in disorganization of the lobular architecture throughout the whole of the liver. Toxic injury to the liver cau ses cellular death through both apoptosis and necrosis. Any cellular injury results in sustained elevation of Ca2+ signaling, which triggers necrotic or apoptotic cellular death [82]. Par ticularly important is an activa tion of nuclear factor kappa B (NFB), which regulates production of cytokines and interferons [83]. Prolonged stress to the cell, particularly the endoplasmic reticulum, can stimulate pro-caspase-12, localized in the endoplasmic reticulum. When activated, caspase 12 stimulates other pro-apoptotic caspases [84]. After cell death, new cells must come in to repl ace the dead ones. Chronic injury results in

PAGE 34

34 cellular regeneration via clonal expa nsion of the resident cells, mainly hepatocytes, which form hepatic foci. They are often referred to as 'regen erative' or 'hyperplastic' foci, implying concepts of pathogenesis rather than a morphological defi nition. They cannot be truly regenerative in that restitution to normal liver tissue does not occur [80]. At the cellular and molecular level, this progressive process is mainly ch aracterized by cellular activation of hepatic stellate cells and aberrant activity of TGF and its downstream cellular mediators (CTGF). The complex signaling pathways of this pivotal cytoki ne during the fibrogenic respon se and its connection to other signal cascades are now understood in some detail. Hepatic stellate cell (HSC) activation is the hallmark of fibrosis. TGF is considered the most powerful mediator of HSC activation in vitro and in vivo [79]. Kupffer cells (KCs) are a main source of TGF in the liver and promote HSC activa tion and fibrogenesis [11; 25]. Since 1989, KCs have been implicated in activating HS Cs [85]. HSC activation and fibrogenesis were almost completely suppressed in KC-deplete d mice with a significan t reduction of hepatic fibrogenesis [25]. Because the tran sdifferentiation of quiescent HSCs to activated myofibroblastlike HSCs is a key event in hepatic fibrogenesis [79; 86], it is likely that the sensitization of quiescent HSCs to TGF and KC induced activation of HSCs constitutes the main mechanism by which inflammation promotes fibrogenesis. HSCs become directly fibrogenic by synthesi zing ECM proteins. In addition, the activated HSC itself proliferates and amplifies the fibr ogenic response. Although th e precise mechanisms responsible for HSC activation remain elusive, substantial insight is being gained into the molecular mechanisms underlying ECM production and cell proliferation in the HSC. The activated HSC becomes responsive to both proliferative (PDGF) and fibrogenic (TGF ) cytokines. These cytokines activate both mitoge n-activated protein kina se (MAPK) signaling,

PAGE 35

35 involving p38, and focal adhesi on kinase signaling cascades. T ogether, these regulate the proliferative response, activating cell cycle pr ogression as well as collagen gene expression. SMAD and p38 MAPK signaling have been found to independently and additively regulate collagen type I gene expression by transcrip tional activation while p38 MAPK, but not SMAD signaling, increases collagen mRNA stability, lead ing to increased synthesis and deposition [87]. Fibroblasts other than HSCs are involved in hepatic fibrosis. The po ssibility of epithelialmesenchymal transition (EMT), which describes the transition of biliary epithelial cells or even hepatocytes to fibroblasts, is st ill under debate. The fibroblasts from EMT actively participate in the generation of fibrotic ECM as established in lung and kidney fibrosis [88]. The role of EMT in hepatic fibrosis remains unknown. The ECM is a complex structure aiding in the support, maintenance a nd regeneration of the liver. This matrix varies throughout the hepatic ar chitecture to correspond to vast functions of this organ. The distribution of the ECM and its associated proteins is important to understand how the liver responds to injury. TGF and CTGF are the key factors for the job. AA and CCl4 are two chronic injury models that demonstrat e this occurrence [89; 90]. These two injury models cause damage to different regions of the liver. Each chemical is metabolized by utilizing molecules specific for their classification. Howeve r they both begin with an oxidation step. Chronic Allyl Alcohol Exposure AA is the smallest representative of allylic alcohols (CH2=CHCH2OH). This chemical is used in industries as a synt hetic intermediate, an effective herbicide and pesticide. AA intoxication leads to periportal (piecemeal) necrosis, ductular proliferation, abundant connective tissue development, macro nodules, cirrhosis, calci um influxes and minimal oval cell infiltration [1; 19; 91; 92; 93]. This is a very useful model for periportal injury because few chemicals cause this type of damage. Some examples of peri portal injuries come from iron overloading from

PAGE 36

36 ferrous sulfate, high levels of exposure from yellow phosphorus, acetic acid and aspirin, and acute viral hepatitis associated w ith cocaine and labetalol [94]. AA oxidation occurs through alcohol dehydr ogenase (ADH). The periportal region possesses higher concentrations of ADH which corre lates with the periportal necrosis seen by AA exposure. AA exerts a dose-dependent toxicity on the cells which is inversely related to cellular glutathione (GSH) [95]. GSH is a ubiquitous tripeptide (including a cysteine) that acts as a powerful antioxidant. It exists in almost every cell of the body, and without it other antioxidants like vitamin E and C are not able to function efficiently [ 19]. AA toxicity can be prevented by inhibitors of ADH and augm ented by the aldehyde dehydrogenase (ALDH) inhibitor disulfiram, which lead to the discove ry that toxicity is produced by a metabolite, specifically the aldehyde acrolein (CH2=CHCH=O) as noted from many references [96; 91; 97]. Acrolein is oxidized by ALDH and has been shown to require NAD+ [98]. This yields free radicals, and protons are transf erred to selenium oxide or alpha carbonyl compounds like ketones which are converted to diketones. ALDH are grouped into classes (ALDH1), and there are around 17 ALDH enzymes in human [99]. ALDH2, 3 and 5 are capable of detoxifying acrolein [100]. Since ALDH2 in rat liver is more concentrated in periportal than in perivenous cells [101] and NAD+ is at a higher concentration around the peri portal regions, this re gion is targeted and damaged. ALDH2 has a low km for acetaldehyde and catalyzes most acetaldehyde oxidation in the liver. It is expressed constitutively, is present exclusively in mitochondria and is NAD+dependent [93]. Acrolein, with ALDH, is an ex cellent substrate for gl utathione-S-transferases [92] and forms adducts with GSH and protein th iols in isolated hepatocytes [95]. In cells depleted of GSH, acrolein may react with e ssential macromolecules and thereby lead to structural and functional derangeme nt and, eventually, irreversible injury. Binding of acrolein to

PAGE 37

37 protein thiols alters secondary and tertiary pr otein structure leading to conformational changes and loss of enzymatic activity [95; 98]. To date, very little data exists on the chroni c exposure of AA. Jung reported that after 16 weeks of twice weekly expos ure (0.62 mmol/kg) severe fibrosis occurs [94]. Chronic Carbon Tetrachloride Exposure Carbon tetrachloride (CCl4) does not occur naturally. It is produced by chlorination of carbon disulfide. It may be a byproduct of dich loromethane and chloroform production as it was first discovered in 1839 by H.V. Regnault, a French chemist and physician. CCl4 is an ozone depleting gas, and has a strong aromatic odor. CCl4 is very stable in air; evaporates quickly and has a half-life of 30-100 years [102]. It has been s hown not to bio-accumulate in animals or soil. It has been used commercially in the producti on of pesticides, solven ts, and antiseptics. CCl4 is primarily metabolized by the membrane bound enzyme cytochrome P4502E1 (CYP2El), which mediates reduction of the toxin [103; 1]. CYP2B1 and B2 are also capable of this action with this chemical [104]. CYP2El is found more prominently on the membranes of the endoplasmic reticulum (ER) [1]. CYP2E1 is constitutively expressed at low levels throughout the body with very high le vels of expression in liver, a nd to a lesser extent in skin, lung and intestine. In the liver, CYP2E1 is concentrated within the centrilobular region, specifically in the first 3-4 layers of hepa tocytes around the central vein [1]. Here, CCl4 is transformed into a trichlor omethyl free radical [1]. In vitro studies indicated that this radical may interact directly with all four DNA bases, with a particular affinity for guanine and adenine [105]. Trichloromethyl free radicals may also react immediately with lipids and cytochromes, leading to a distinct, locally necrosis. Trichl oromethyl free radical can also undergo anaerobic reactions, which may result in the formati on of such toxic compounds as chloroform,

PAGE 38

38 hexachloroethane, and carbon monoxide [1]. Th is radical reacts with GSH to form GSHcontaining radicals [106] and has been consid ered to directly act on hepatocytes since 1973 [107]. This formation causes lipid peroxidation in the ER, damage to the plasma membrane and an increase to intracellular calc ium content [108]. Lipid free radi cals and peroxides fare less reactive, and diffuse out to cells throughout th e lobule. These radicals can, however, cause oxidative damage to cell membrane s and ER. Lipid peroxidation also results in the production of aldehydes which require metabolism by ALDH to prevent adduct formation with cellular macromolecules [108; 109; 110]. Smuckler in 1976 confirmed that CCl4 induces steatosis by disruption of lipid transport. Synthesis of fat was not increased. He also repor ted that decreased lipid transport was due to problems with protein synthesis based on inad equate ribosomal activity. Many hepatocyte proteins, including albumin, are essential for tr ansporting lipids out of a cell are now reduced leading to more downs tream effects [111]. In 1936 Cameron reported that in acute studies, CCl4 has been shown repeatedly to cause massive necrosis of the central vein within 24hr of exposure fo llowed by rapid proliferation of the remaining cells within 5-6d to reach full re storation if it was a one time exposure [111]. Chronic exposure leads to a much more serious problem. Repeated exposures lead to more central vein damage and necrosis, and subseque ntial periportal damage, micro nodules, excessive lipid build up (steatosis), a bundant connective tissue develo pment, disruption in calcium homeostasis, streaming cords of fibrous tissue, cirrhosis, oval cell and bili ary cell infiltration and has significant risk of cancer [1; 19; 111; 112]. In rats treated twice weekly with CCl4 for four weeks, significant liver fibrosis with septae formation occurred. When CCl4 treatment was terminated at four weeks, the liver fibrosis resolves completely and normal liver histology was

PAGE 39

39 restored. Interestingly, upon termination of CCl4 treatment, 50% of the activated HSCs underwent apoptosis with an accompanying increase in MMPs and decrease in TIMPs [24]. In another study by Ruchirawat acute CCl4 was shown to cause a bloc kage of DNA methylation, and over ten years later it was shown by Vare la-Morieras that hypomethylation occurs with chronic exposure, leading to the disruption of the careful balanci ng act of gene regulation [113; 114]. Degradation of Pre-Existing Matrix Current evidence indicates that matrix degrad ation is of central importance for hepatic fibrogenesis [15; 116], but it is also involved in liver regeneration and carcinogenesis [15; 117; 118]. Matrix metalloproteinases (M MPs) are the main enzymes that degrade ECM proteins. They have a broad substrate specificity which de grades collagens, laminins, fibronectins, proteoglycans as well as other matrix compone nts. Expression of MMPs in hepatocytes and mesenchymal cells may be mediated by the transcript ion factor AP-1. This tr anscription factor is upregulated by many pathol ogical stimuli, including inflammatory cytokines, growth factors and other stress signals. AP-1 repr esents a single component of a complex, dynamic network of signaling pathways involved in the response to hepa tic injury, so it is like ly that several other factors co-regulate MMP expression [119]. MMPs are secreted from cells into the ex tracellular space as proenzymes, which are subsequently activated by HSC cell surface-associated proteases. The active enzymes are, in turn, inhibited by a family of TIMPs. This combination of positive and negative regulation governs matrix degradation. Additional matrix pr oteases have been implicated in liver ECM remodeling. Progelatinase A is secreted by activ ated HSCs in culture. This proenzyme is activated by a protein complex found within the HSC membrane The activation of matrix proteases at the HSC membrane indicate that the active HSC is able to independently remodel

PAGE 40

40 matrix by breaking down existing ECM as new ECM is produced [19]. De gradation of existing matrix does not necessarily limit fibrosis, as numerous growth factors adhered to the extracellular matrix are liberated, creating a new and progressive microenvironment through which the fibrogenic message is passed [120]. The balance between th ese two competing HSC functions determines whethe r or not fibrosis occurs. Overview of the Transforming Growth Factor Beta There are over 100 distin ct members of the TGF superfamily sharing at least one conserved amino acid domain [121]. This prot ein was discovered by Roberts in 1983 while conducting studies on fibrobl asts that had been re trovirally transformed in vitro This growth factor has been described as bot h heat and acid stable, and may be extracted from a wide variety of cells [122]. TGF is a 25 kD cytokine that inhi bits epithelial cell and lymphocyte proliferation. Conversely, it is able to stimulate pro liferation and extracellular matrix production by fibroblasts [89; 123]. TGF is also known to play a key ro le in tissue remodeling and in immunological self-tolerance [124]. In addition to being directly secreted by cells, a reservoir of TGF is covalently bound to the N-te rminus of collagen IIA [125]. Significant production of this cytokine occurs within hepatic fibrobl asts and macrophages, and represents a major component of the secret ory vesicles of platelets. Work by Carr in 1986 showed that TGF is inhibited during severe liver injur y, allowing cell prolif eration. In cultures of primary rat hepatocytes, TGF bound its receptor with an apparent Kd of 93.1 pM. This binding constant was found to be transiently dimini shed following partial hepatectomy, returning to normal by 96 hours [126]. Three members of the TGF family (TGF 1, 2 and 3) have been identified in mammals. TGF 4 and 5 are found in the chicken and Xenopus respectively. TGF requires proteolytic cleavage to an active form before it will bind to its receptor. Cleav age also mediates the release

PAGE 41

41 of extracellular TGF from the ECM. To date, five TGF receptors have been identified (TGF R1-5). TGF R1 is expressed predominantly by he matopoietic progenitor cells, where it has been shown to form a heterotetramer with TGF R2 [127]. Activation of Transforming Growth Factor Beta As mentioned previously, TGF is synthesized in a latent precursor form that is subsequently cleaved to the 112 am ino acid active form [128]. TGF 1 is considered to be the predominant isoform of the molecule in mo st cell types and is highly conserved among mammalian species. Unless specified otherwise, TGF will refer to TGF type 1 throughout the remainder of this report. N-terminus cleavage of the TGF pro-peptide occurs within the Golgi apparatus. The truncated C-termini dimerize via disulphide bond formation between conserved cysteine residues, giving rise to the s ecretable, latent form [129]. Fu rther cleavage by extracellular proteases generates the biologically activ e form of the molecule. Activated TGF readily forms a complex with 2 macroglobulin in serum, and is rapidl y cleared by the liver resulting in a t1/2 of only 3 minutes [130]. The latent form does not complex with 2 macroglobulin and is able to persist for a much longer time in serum [131]. TGF displays a high affinity for the type II receptor [132]. TGF binding to the type II receptor mediates the formation of a heterotetramer composed of two type I and type II subunits. Constitutively active kinase activity associated w ith the type II receptor trans-phosphorylates specific serine residues on the type I receptor, initiating an intracellula r signaling cascade which is propagated to the nucleus [132]. The activated TGF receptor tetramer phosphorylates members of the Smad protein family, likely Smad 2 and 3. These proteins form a complex with Smad 4 within the nucleus and recruit additional proteins that form a transcri ption complex that initiat es gene transcription.

PAGE 42

42 There is evidence that the TGF R tetramer is rapidly endocytosed via a clathrin dependant process. However, this interna lization is not required for Sm ad phosphorylation and is likely a mechanism for receptor recycling [127; 132]. Biological Activity of Transf orming Growth Factor Beta TGF has been shown to elicit multiple biologi cal effects dependant on cell type and the presence of other growth factors. It can either stimulate or inhibit cell proliferation and may regulate the action of other growth factors [133]. TGF stimulates the synthesis of major extracellular matrix proteins, in cluding collagen, proteoglycans, glycosaminoglycans, fibronectin and integrins. TGF also facilitates ECM deposition by in hibiting the synthesis of matrix metalloproteinases (MMPs) and i nducing the production of tissue protease inhibitors (TIMPs). This influence on protease activity is sp ecies, tissue and cell type specific. TGF has also been demonstrated to play a key role in carcinogenesi s, as it is able to re gulate expression of both proto-onco genes and tumor suppressor genes. TGF has also been shown to function through bo th the mitogen-activated protein kinase (MAPK) and the phosphatidyli nositol 3-kinase (PI3K) signa ling cascades. Together, these regulate proliferative responses, activating cell cycle progression as well as collagen gene expression. Smad and MAPK signaling have be en found to independently and additively regulate collagen I gene expression while MAPK but not Smad signaling increases collagen I mRNA stability [87]. With respect to liver fibrosis, TGF mediated activation of Smad 2/3, appears to be the key signaling pathway [124]. Overexpression of TGF in rodent models yields an increased expression of protease inhibitors, such as TIMP and PAI-1 resulting in excessive matrix accumulation. This increase in protease inhibition coupled with an increased rate of matrix deposition forms the cornerstone of fibrosis [120]. TGF has been reported to reduce the production of collagenases MMP-1 and

PAGE 43

43 stromelysin MMP-3, but enhance th e expression of the inhibitors TIMP-1 and TIMP-3 in human lung fibroblasts, myometrial smooth muscle cel ls and articular chondrocytes [134; 135]. In addition, the production of several matrix compon ents, such as collagen and fibronectin, is induced by TGF 1 [125]. Latent TGF is rapidly taken up by the liver cells including hepatocytes [131; 136]. High levels of TGF within the hepatocytes can lead to apoptosis [136]. When exogenous TGF is administered following 70% partial hepatectomy, hepatocytes and stellate cells were inhibited from proliferating, but not the e ndothelial cells [137]. This is consistent with another report demonstrating that TGF can act synergistic ally with endothelial growth factor (EGF) to either stimulate or inhibit endot helial cell colony formation in vitro depending on concentration. TGF treated livers eventually rec over from partial hepatectomy, as hepatocytes overcome inhibition and proliferate. Recently, TGF has been shown to stimulate CTGF expression in hepatocytes [138; Gressner OA, 2007]. TGF can be inhibited by addition of TGF RII [139]. Overview of Connective Tissue Growth Factor During tissue repair and early development, TGF gene expression is coordinately regulated with connective tissue growth factor (CTGF) [140]. This is putatively due to the TGF -responsive element located within the CTGF prom oter [141]. With respect to signaling molecules, SMADs, ras/MEK/ERK, PKC ar e required during CTGF induction by TGF in cultured mesangial cells [142]. CTGF is a cysteine-rich secr etory protein of 36-38 kDa, that belongs to the CCN protein family, named after C TGF, c ysteine-rich 61, and n ephroblastoma overexpressed proteins [143]. All members of th is family exhibit a high degree of amino acid sequence homology (50-90%) possess a secretory si gnal peptide at their N-terminus and contain four distinct protein modules. Module 1 is homologous to the N-terminal cysteine-rich regions of the six "classic" insulin like growth factor binding proteins (IGFBP-1 to -6) [143], and contains a

PAGE 44

44 motif (GCGCCXXC) that is involved in binding insulin like growth factor (IGF) with low affinity [144]. The physiological ro le of this binding site still remains to be defined. Module 2 contains a von Willebrand type C domain (vWC) that occurs in von Willebrand factor (vWF) as well as various mucins, thrombospondins, and collagens [143] and may be involved in oligomerization [145]. Module 3 is a thrombospondin type 1 domain (TSP1) that contains the local motif WSXCSXXCG and appears to be a cell attachment factor that binds sulfated glycoconjugates [143; 146]. Module 4 is a C-terminal module that also occurs in the C-termini of a variety of unrelat ed soluble proteins including TGF and PDGF [143] and is responsible for dimerization and receptor-binding [143]. The bio active form of porcine CTGF is contained within the 103 C-terminal residues of the primary translational product [147] and thus supports the proposed role of module 4 in binding cell surface receptors [143]. Pi et al re ported that the Cand N-terminal ends of CTGF are able to bind fibronectin [148]. Biological Significance of Conn ective Tissue Growth Factor CTGF is associated with immediate early grow th responsive genes that putatively regulate the proliferation/differentiati on of various connec tive tissue cell types during embryogenesis [147]. It is a cysteine-rich, matrix-associated, heparin, perlacan, fibronectin and integrin-binding protein that promotes endothe lial cell growth, migration, adhesion and survival, while stimulating ECM production [149], chemotaxis, pro liferation and angiogenesis. However, CTGF can act as cell growth inhib itor and induce apoptosis. In so me cases, it is intrinsically nonmitogenic and augments the activity of othe r growth factors [150]. In addition, CTGF was shown to be positively regulated by vascular endothelia l growth factor, epidermal growth factor, fibroblast growth factor [151], plasma clotting factor VI Ia, thrombin [152; 153] and by lysophosphatidic acid and serotonin activation of heptahelical receptors [154] but negatively regulated by tumor necrosis factor[155].

PAGE 45

45 CTGF was first identified in conditioned media containing human umbilical vein endothelial cells in culture and was found to be mitogenic and chemotactic to cells of connective tissues such as fibroblasts [156] Besides fibroblasts, it is expr essed in, and acts on endothelial cells, skeletal and smooth muscle cells, chondr ocytes, epithelial cells, neural cells and hepatocytes. It has also been implicated in several normal physiological processes, including those related to embryo development and diffe rentiation [150], endochondr al ossification [157], and female reproductive tract function in the ut erus and ovary [158]. Up regulation of CTGF has been linked to many pathogeneses including fi brosis [159], tumor de smoplasia [160], wound healing, and tissue regeneration. A rodent model with CTGF knocked out showed impaired chondrocyte proliferation, angiogenesis, ECM production and turnover and abnormal skeletal growth [161]. In liver, CTGF is normally expressed in a very low level, except for a local upregulation in portal and central vein endothelia and in myocytes of portal arteri es [27]. However, a significant induction of CTGF transcripts has been found in liver injury [162] and pathogenesis including liver fibrosis in animal mode ls and in human chronic liver diseases [Paradis, 1999; 164]. Activated stellate cells are f ound to be a major cellular source of CTGF in fibrotic and CCl4injuried rat livers. When added in primary stella te cell cultures, CTGF strongly promote stellate cell proliferation, migration and produce collagen type I [163]. Recently, proliferating epithelial cells in bile duct and in hepatocytes have also been shown to produce CTGF mRNA by studying the temporospatial expression of CTGF in rats with acute and ch ronic hepatic fibrogenesis [27; 88] CTGF is known to induce phosphorylation of p42/44 MAPK and protein kinase B in primary mesangial cells via a integrin dependent manner [165]. Th ese signaling pathways are likely the mechanism through which CTGF induces phenotype changes in target cells. The multiligand

PAGE 46

46 receptor, low density lipoprotein receptor (LRP)-related protein/2-macroglobulin receptor is a protein complex that binds CTGF and mediates internalization and degradation to negatively regulate serum CTGF levels [166]. During liver fibrosis, CTGF expression is strongly up-regulated. This increase may be easily measured in serum. This finding is corrobora ted by the fact that CTGF levels decrease in fully developed, end stage cirrhosis, when the fibrotic process is complete [88]. Specific Aims of this Study Chronic chemical injuries occur throughout the world as medical and industrial environments continue to expand. The liver is re sponsible for the metabolis m of the majority of these chemicals. The resulting metabolites may e ither be useful or toxic by-products. Allyl alcohol and carbon tetrachloride ar e chemicals which affect the ge neral public in a variety of ways. These chemicals injure different zones of the liver. The goal of this research was to investigate differences in the fibrotic response to chronic hepatic portal zone injury versus chronic central zone injury. This lead to th e desire to understand how the ECM is laid down during these chronic liver injuries. For this st udy we specifically asked how can protein and nucleic acids be extracted from archival forma lin fixed paraffin embedded (FFPE) fibrotic liver? Is a periportal injury more seve re than a pericentral injury? Where are these proteins expressed, and what is the composition of these proteins in the ECM? To begin answering these questions, experiments were performed to: optimize extractions from fresh, archival and chronic injury archival FFPE tissue characterize fresh, archival and chronic injury archival FFPE tissue through IHC This study incorporated methods from molecu lar genetics, molecular biology, toxicology, and pathology.

PAGE 47

47 Figure 1-1 Hepatic microarchitect ure and blood flow. Red arrows indicate blood flow and green arrows indicate the direction of bile fl ow. M. H. Ross All rights reserved. Figure 1-2 Liver acinus. The three zones of the liver radiating toward the central veins. M. H. Ross All rights reserved. Figure 1-3 Growth of remain ing three liver lobes after partial hepatectomy in the rat. Compensatory hyperplasia results in the liver regaining original tissue mass in approximately 10 to 14 days.6 Amer ican Medical Asso ciation. All Rights Reserved. Figure 1-4 Amount of resident hepatic cells within the cell cycle during the time following partial hepatectomy. Hepatocytes represent the proliferative driving source behind liver regeneration. AAAS. All Rights Reserved.

PAGE 48

48 CHAPTER 2 MATERIALS AND METHODS Experimental Animals Male Fisher-344 rats (Fredricks Laboratories, Fredrick, MD) were used for all experiments described. The dosage for chronic exposure to eith er allyl alcohol (AA) or carbon tetrachloride (CCl4) (Sigma, St. Louis, MO) was determined mathematically calculating the LD10 dose. Chronic dose of AA consisted of 0.001 mL/kg ( 7.4 mg/kg) body weight in a dilution of 1:50 vol/vol in 0.9% saline solution. Chronic dose of CCl4 was given as 0.38 mL/kg (300 mg/kg) of body weight in a 1:1 vol/vol dilu tion in corn oil. These chemi cals were administered by i.p. injection using a pattern of five consecutive days of injections followed by two days of rest for a duration of 90 days. They were given this treatm ent at the same time of the day for everyday treated. This exposure pattern w ould closely mimic an average wo rker's exposure during a threemonth period. The animals were sacrificed 5-6hr after the final inject ion. Two hours prior to sacrifice, animals were injected with BrdU (50 mg/kg body weight) for determination of DNA synthesis as described by Lindroos et al [50]. All procedures regarding animals were conducted according to institutionally approved protocols. Animal Sacrifice and Tissue Collection All animals utilized for tissue collection were euthanized by administration of an overdose of pentobarbital (1.0 mL/100g body weight, Si gma). This is consistent with the recommendations of the panel on euthanasia of the American Veterinary Medical Association and the Guide for the Use and Care of Laborat ory Animals (U.S. Department of Health and Human Services/NIH Publication #8 6-23). After euthanasia, sections of liver were collected for paraffin embedding. These sections were fixed O/N in 10% Neutral Buffered Formalin (RichardAllan Scientific, Kalamazoo, MI). The formalin was then exchanged for PBS and embedded in

PAGE 49

49 paraffin. All paraffin sec tions were cut to a 5 m thickness. Liver sections were collected at the following time points: day 15, 30, 40, 45, 55, 60, 65, 70 and 90. Protein Isolation, Purifica tion and Quantification Fresh Normal Rat Liver Extraction Protein was isolated from formalin fixe d paraffin embedded (FFPE) rat liver tissue. Controls used were snap frozen normal rat li ver (NRL) tissue, fresh FFPE NRL and archival FFPE NRL. Whole liver sections from snap fr ozen NRL were cut into small pieces (about 150200g) and added to 10ml of RIPA buffer (50mM Tris-Cl pH 7.4, 250mM NaCl, 5mM EDTA, 50mM NaF) with proteinase inhibitors (per 1mL of RIPA buffer 30l Aprotinin, 10L 1mM leupeptin, 10L 100mM phenylmethylsulfonyl fl uoride (PMSF) and 10L Na orthovandate were added). The fresh tissue is kept on ice for th e remainder of the proce dures until it is ready for analysis. The fresh tissue is disrupted utilizing either a homogenizer (10sec at a time until the proper consistency is reached) or a needle (18-26.5 gauge) and syringe where it is pipetted up and down until tissue was thoroughly homogenized. The sample was vortexed for 30sec and then centrifuged at 10,000xg at 4C for 10min to remove excess lipids and DNA. The supernatant was collected into 2.0mL screw cap t ube and place in -80C until use. Formalin Fixed Paraffin Embedded Tissue Extraction For FFPE tissues, preceding methods were necessary due to paraffin and formalin contaminations. The following was performed individually and in different combinations. Tissues were either cut on a microtome or a porti on of the liver section was dissected out with a scalpel. They were then subjec ted to paraffin removal via heat and or xylene treatments. The buffer solution (RIPA and proteinase inhibito rs) was altered through chemical composition, protease inhibitors and pH. More stringent di sruption methods were added. Cleanup was also

PAGE 50

50 necessary and performed through acetone precip itation(s). Samples were analyzed through a modified Lowry assay from Bio-Rad (DC pr otein assay, Bio-Rad, Hercules, CA). Amounts and sizes of tissues were varied by cutting the samples on a microtome from 540m thick, yielding 5 to 500m total, or by re moving a piece with a scalpel from about 150200g. This piece is subjected to ei ther going straight into the next step or is broken down into smaller pieces with a mallet. Tissues were heat ed to varying times and temperatures to allow paraffin to melt and assist in breaking the cros slinking. Heating was performed on aluminum foil upon a hot plate or inside a plastic heat sealed bag in boiling or near boiling (90C) water. Paraffin removal through xylen e was performed for 5min x 2 to 30min x 2 followed by dehydration through 100% ethanol (EtOH) 5min then 30 min, 90% EtOH for 5min then 30min, 70% EtOH for 5min and then into a buffer solutio n. In the buffer solution, the tissue was either transferred into a plastic bag a nd disrupted with a mallet, or in to a tube to be homogenized as described for normal tissue. Buffer solutions were altered through a variet y of techniques. Alteri ng the pH was the first attempt, ranging from a very acidic 2.0 to a hi ghly caustic 13.4. The next option in altering the buffer was changing the percent of chemicals such as SDS from 0% up to 2% (typical is 0.1%) and EDTA (from 1mM to 10mM with 5mM being standard). Citrate buffer was added to the solution to aid in the remova l of protein crosslinking. Then protein handling was considered. The prot eins typically are used the same day as extraction and are not frozen. Freeze/thaw cyc ling was performed from one to three times. Boiling was also performed at varying times (0-2 0min) in the varying solutions for crosslink removal.

PAGE 51

51 Acetone Precipitation The appropriate amount of acetone was cooled to -20C and added in four times the volume of the protein/RIPA buffe r solution. The samples were eith er vortexed or inverted then incubated at -20C for 1hr to overnight, and th en were centrifuged for 10min at 14000xg at 4C. The supernatant was saved for further extracti on. The pellet was allowed to dry at RT for 1025min, not allowing it to over dry. Buffer is added to reconstitute the protein in the appropriate amounts depending on the pellet size and subseque nt spectrophotometer reading (DC protein assay). This step was repeated as necessary. Protein Quantification with DC Protein Assay Blank and protein standards were made in 1ml tube with 1uL sample and 24 L RIPA Buffer without Protease Inhibito r Solution in 1ml clear tube. In another 1ml tube 125L Reagent A per reaction and 2.5L Reagent S per reaction from the DC Protein Assay (Bio-Rad) were mixed. Note: Reaction number equals Sample numb er plus 5. 125L of combined solutions A& S was added to each reaction. When ready to measure 1ml Reagent B was added and tubes vortexed. 5-10min after the addi tion of solution B the OD of the samples were measured in disposable cuvettes in Spect rophotometer set to 750nm. Running Gel The sponge on gel pouring apparatus was dampen ed and the plates cleaned with alcohol. The plates were then aligned and secured in ge l pouring apparatus. The gels were poured using electronic or plastic Pasteur pi pettes. They were poured to approximately 1-1.25cm below the top of the plate. Butyl alcohol was added to t op of glass to prevent bubbles and smiling of gel. After 30min the remaining solution in the 15ml tube was inverted to determine if gel solidified. The butyl alcohol was removed and the gel rinsed with Milli-Q H2O until the odor dissipated.

PAGE 52

52 Stacking Gel The gels were pored using elec tronic or plastic Pasteur pipette s to slightly below top of glass and the comb inserted. After 10-15min, th e gel was ready to run after gently removing comb and rinsing the wells with Milli-Q H2O. Western Blot Analysis Protein Sample Preparation The amount of protein to be loaded per well was determined based on the source of isolation, archival or fresh FFPE, and the sensi tivity of the antibody being used for detection. Samples were added to the appropriate amount of buffer and placed in a screw cap 2.0mL tube. 5L of 5X Western Loading Buffer per lane was a dded to each tube. Each sample was boiled for 0-20min and then incubated at RT for 5min to cool. The samples were immediately loaded and any remaining solution placed on ice and returned to storage at -80C. Each well of 0.75-1.5mm gel was loaded with 15-40L. Electrophoresis of the Western Gel The gel was loaded in the r unning apparatus with small pl ate facing inward. The inner chamber was filled with 1X R unning Buffer till full and over fl ows and the outer chamber was filled 2.0 inches. 15-40 L of samples was loaded per well with 3 L of Protein Standard (kaleidoscope, Bio-Rad) within the appropriate well. Any empt y lanes were filled with 5 l of 5X Western Loading Buffer. The gel was run at 60-80 Volts (based on individual set up flow rates of the bubbles) until the loading die had migrated out of the stacking gel. Then the gel was run at 90 Volts until the loading die ran the length of the gel. Transferring of a Western Gel to a PVDF Membrane The upper left corner of the Immuno-Blot PVDF (Bio-Rad) membrane was cut and the membrane was labeled with pencil It was then dipped in methanol and soaked in water for five

PAGE 53

53 minutes then in 1X transfer buffer for 20min. S ponges and filter papers were also soaked in transfer buffer. The gel plates were opened and the stacking gel/wells were removed. The gel was submerged in 1X transfer buffer. A sandw ich consisting of black assembly tray, sponge, filter paper, gel, PVDF membrane filter paper, sponge, and red assembly tray along with an ice block and stir bar were placed in the transfer a pparatus. The transfer apparatus was filled with 1X transfer buffer and placed on a stir plate. The transfer buffe r was stirred continuously while transferring to ensure the appa ratus would not overheat. The prot eins were transferred at 150 milliamps for 60min for a 75mm gel and 90min for a 150mm gel. Coomassie Staining of Western Gel In order to determine protein banding patterns or efficiency of transfers, the gel was placed in 20mL of methanol, 7.5mL of glacial acetic acid, and 72.5mL of Milli-Q H2O and shaken at RT for 30min. The gel was stained with Coom assie (0.4g of Coomassie blue R350 (Sigma), 80mL methanol, 40mL glacial acetic acid, and 280mL Milli-Q H2O) for 30min at RT. It was then destained as desired in 500ml methanol, 100ml glacial acetic acid, and 400mL Milli-Q H2O. A Kimwipe (Kimberly-Clarke, Roswell, GA) was placed in the solution a nd as it was saturated with color it was replaced until the desi red level of destaining was achieved. Silver Staining of Western Gel The SilverQuest Silver Staining Kit (I nvitrogen, Carlsbad, CA) provides a more sensitive detection of proteins than the Coomassie stain. All solutions were made at the time of use. First the gel was soaked in 7% acetic acid for 7min, transferred to 200mL of 50% methanol for 20min X 2, rinsed in 200mL wa ter for 10min X 2, and 5min before the end of the final water rinse the staining solution was made. The gel was then soaked in the staining solution for 15min, rinsed in 200mL of water 5min X 2. The gel wa s soaked in developing solution until bands were visible, about 2 to 15min then stopped by rinsing in water 3 ch anges of water. Cellophane was

PAGE 54

54 then wrapped around the gel and affixed on a glass pl ate that is larger th an the gel, the bubbles were removed and it was allowed to dry ove rnight before it was cut to size. Probing of Western Membrane The membrane was blocked for 1hr at RT or overnight with a bloc king solution consisting of 5g skim milk and 100mL 1X PBS, then wa shed with PBS-T for 5min. The membrane was probed with the appropriate conc entration of primary antibody overnight at 4C. The membrane was then rinsed 3x for 5min each with 1X PBS. The appropriate horseradish peroxidase conjugated secondary antibody was applied in 1X PBS for 30min to 1hr shaking at RT. The membrane was then rinsed again 3X with 1X PBS. Developing of Western Membrane with ECL Plus Excess liquid was removed from the membrane and it was placed within a plastic bag. 25 L of Solution A mixed with 1mL of Soluti on B ECL Plus reagents (GE Healthcare, Piscataway, NJ) was incubated on the membra ne for 5min. Excess ECL Plus reagent was removed. Film was exposed to the membrane fo r 5sec to overnight depending on the brightness of the banding pattern. The membrane was then stripped if further probing was necessary. Membrane Stripping for Reprobing 20mL of 5X stripping soluti on was diluted to 1X with 80m l water (100ml total). Then 714 L -Mercaptoethanol was added and membrane placed within the solution. The membranes were incubated in closed Tupperware contai ner at 56C for no longe r than 30min with intermittent shaking. The membrane was then wa shed for 5-8 X 5min each with 1X PBS-T until all residual B-Mercaptoethanol was removed. Membranes were then reblocked with milk and reprobed as mentioned.

PAGE 55

55 Nucleic Acid Isolation, Purifi cation and Quantification RNA Isolation and Analysis RNA isolation methods included optimizi ng crude extraction th rough a homogenizer, syringe, grinder and/or mallet; varying pH (2-13.4) ; heat extractions (melting sections on a hot plate to varying times and/or va rying boiling times), changing th e percent of chemicals like SDS from 0.1% up to 2% and EDTA. RN Azol Bee is used by adding 20 L of the RNAzol Bee solution, 80 L of RNAse free water, 350 L RLT/Beta-Mercaptinoethanol and 250 L 100% EtOH to 30-50 L of RNA solution. Then inverting the solution before pipetting 700 L into the Qiagen Minispin cleanup pink columns. The columns were then centrifuged for 15sec at 10,000RPM and the collection tube was emptied, 500 L of Buffer RPE was added then centrifuged for 15sec at 10,000RPM and repeated for 2min. Excess buffer was removed by full speed centrifugation for 1min. A new tube was used to elute with of RNase-free water in the amount according to the pellet size (10-50 L). Then directly frozen at -80C or put in 55C water bath for 10min and inverted then either used or froze at -80C, or inverted at room temperature and the concentrations were determined by sp ectrophotometry. Another method used was a kit from Ambion, RecoverallTM Total Nucleic Acid Isolation (Amb ion, Austin, TX). This kit also did not yield any produc t for RNA or DNA. Reverse Transcription PCR RNA was converted into c DNA through reverse transcrip tion PCR (rtPCR) in the following procedure. 1-5 g of RNA was added to a mixture of 10mm dNTPs (1 L per 20 l reaction), random hexamers (2.5 L per 20 L reaction) and water then were incubated at 65C for 5min then ice for 1min. Then 5 L of 10x RT buffer, 0.1x DDT and RNase Out are added in a 2:2:1 respectfully. The samples were mixe d and incubated at 42C for 2min and 1 L of

PAGE 56

56 Superscript II was added to each and continued at 42C for another 50min until terminated at 70C for 15min. 1 L of RNase H was added and incubated at 37C for 20min to rid the samples from any residual RNA. All products are from I nvitrogen's Superscript II kit. The cDNA was processed as DNA as listed in PCR Analysis. DNA Isolation and Analysis Altered methods utilized included optimiz ing crude extraction through a homogenizer, syringe, grinder and/or mallet; varying pH (2-13.4) ; heat extractions (melting sections on a hot plate to varying times and/or va rying boiling times), changing th e percent of chemicals like SDS from 0.1% up to 2% and EDTA. The main issue is contamination of cellula r debris and formalin cross-linking, so the tissues were isolated and washed multiple times. Isolation and washing were also varied. Samples are reisolated through Pr omega's DNA Wizzard, Sigma's PCI mix, or plain phenol, then are mixed via vortexing or inver ting for a range of times (1min-4hrs) then centrifuged at 14000xg for 10-15min at 4C. The t op clear supernatant is removed with or without the white protein layer while discarding the bottom phenol layer. Either isopropanol (at 1:1) or 100% ethanol (1: 2.5) is added with 1/10th volume of 3M sodium acetate pH 5.2 to visualize DNA in solution or pelleting. This continued until the protein layer was no longer visible which decreased the c oncentration, but more effici ently purified the DNA for PCR analysis. PCR Analysis DNA concentrations were measured with a spectrophotometer but they were not completely accurate due to the contamination pr oblems faced with utiliz ing fibrotic, archival formalin fixed paraffin embedded tissues. A subs tantial amount of DNA was lost with multiple isolations and washings. P CR was performed with 0.5-5 g of DNA, GAPDH primers at varying

PAGE 57

57 concentrations of .25-1 L, dNTPs at 0.6 L, 2 L Eppendorph's smart taq buffer with magnesium balance, 0.2 l Fisher brand Taq in a 20 L reaction. GAPDH primers were as followed: Forward: ATCACTGCCACTCAGAAGAC; Reverse: AACACTGAGCATCTCCCTCA. The PCR was performed at a melting temperature of 95C fo r 5min followed by 25-40 cycles of 95C for 30sec, 58C for 1min and 72C for 1-1.5min and finished by 10min at 72C. The samples were cooled to RT or stored in 4C before being loaded into a 1.5-2 % agarose gel with 1% ethidium bromide. Histological Analysis Hematoxylin and Eosin Staining of Paraffin Embedded Tissue Tissue sections of 5 M in size were cut and placed in a 42C water bath. There were then lifted from the bath with a Superfrost Plus (Thermo Fisher Scientif ic Inc., Waltham, MA) positively charged slide. The slides were air dr ied overnight at RT. The paraffin was removed and the slides rehydrated by incubating them in Xylene 2 x 5min, 100% ethanol 2 x 2min, 95% x ethanol 2 x 1min, and distilled H2O for 1min. If an antigen retrie val step is necessary, it is performed at this point then rinsed with dist illed water before proceeding. Nucleic acids and other positively charged molecu les were then stained with Hematoxylin 7211 (Richard-Allan Scientific, Kalamazoo, MI) for 2min 15sec and rinsed with distilled H2O for 2 x 1min. The blue color of the Hematoxylin was inte nsified by incubating th e slides in Clarifie r 1 (Richard-Allan Scientific) for 1min, distilled H2O for 1min, Bluing Reagent (Richard-Allan Scientific) for 1min, distilled H2O for 1min, and 80% EtOH for 1min. Proteins were then stained a pink color with Eosin-Y (Richard-Allan Scientific) for 1min 30sec. The tissue was then dehydrated for coverslipping with 2 x 1min 95% EtOH, 2 x 1m in 100% EtOH, and 3 x 1min Xylene. Coverslips were then applied with Cytoseal XYL (Richard-Allan Scientific).

PAGE 58

58 Gomori's Trichrome Tissue sections of 5 M in size were cut and placed in a 42C water bath. There were then lifted from the bath with a Superfrost Plus posi tively charged slide. The slides were air dried O/N at RT. The sections were then put into Xylene 5min x 2; 100% EtOH 2min x 2; 95% EtOH 3min; H2O 1min; Bouin's (preheated) 1hr 56, rinse in running tap H2O 3-8min until yellow color is removed; Hematoxylin 10min; Rinse in running tap H2O 5min; Trichrome 15min; 1% Acetic Acid 1min; Ri nse in distilled H2O 30sec; 100% EtOH 1min x 2; Xylene 1min x 3; Cytoseal/Coverslip (Trichrome kit: Richard Allan: 87021). Chromagen Staining Tissue sections of 5 M in size were cut and placed in a 42C water bath. There were then lifted from the bath with a Superfrost Plus posi tively charged slide. The slides were air dried O/N at RT. The paraffin was removed and the slid es rehydrated by incubati ng them in Xylene 2 x 5min, 100% Ethanol 2 x 2min, 95% EtOH 2 x 1min, and distilled H2O for 1min. If an antigen retrieval step is necessary, it is performed at this point then rins ed with distilled water before proceeding. The slides were then transferred to a slide cassette holder and rinsed with TBS-T 5min. Liver contains endogenous bi otin, biotin receptors and avidin binding sites and in order to utilize The Vectastain Elite kit (a very sensitive procedure), th ey must be blocked to prevent nonspecific staining. Another source of nonspeci fic staining can occur with the secondary antibody when this kit can bind to the antigens of the antibody source. Serum blocking was utilized by taking the serum of the animal that the secondary antibody wa s made. To conserve time and the avidin and biotin solutions, they were diluted by perfor ming two serum blocking steps. Avidin (15ul/ml of 1mL TBS-T with 4 drops avidin) 15min and washed in TBS-T 5min, then biotin (15uL/mL of 1mL TB S-T with 4 drops biotin) 15mi n and washed in TBS-T 5min. The primary antibody is added to the optimum con centration (as listed in the following table)

PAGE 59

59 diluted in an antibody diluent (Invitrogen, Carlsba d, CA), covered (polyethylene class 4, cut just smaller than the size of the slide) and kept ove rnight at 4C. The slides were uncovered and washed with TBS-T 5min then the secondary antibody (1 :200) is added 30min. Then washed in TBS-T 5min, ABC 30 min, wash in TBS-T 5 mi n and then DAB (in 2.5mL tap water 1 drop buffer, 1 drop H2O2, 2 drops DAB) to optimize; H2O 1-5 min; Hematoxylin to optimize; H20 15sec x 1; 1min x 1; Clarifying solution if necessary up to one min. Bluing optimize; H20 1min; 70% EtOH 1min; 95% EtOH 1min; 100% EtOH 1min x 2; Xylene 1 min x 3; Cytoseal XYL and coverslipped. All chromagen stains were run with ma tching Immunoglobulin cont rols. Avidin/biotin solutions (SP-2001), normal serum (Goat, Horse a nd Mouse), secondary antibodies (goat, rabbit and mouse), ABC kit (PK-7100), DAB kit (SK-4100) we re utilized at room temperature and are from Vector Laboratories, Burlingame, CA TBS-T is 1X TBS with 0.1% Tween-20. All samples were photographed using an Ol ympus microscope and Optronics digital camera (Olympus, Melville, NY) Antigen Retrieval Trypsin digest : 10min 37 (Lab Vision: AP-9008-3A and -1B, Fremont, CA) rinse with water and resume. Proteinase K digest : 10min 37 (Dako: S3020) ri nse with water and resume. Citra retrieval : pH 6.0, 1x Sodium Citrate in H20; 7min in a microwave at 50% power, 18min cooling (Sigma: S1804), rinse with water and resume. Immunohistochemistry Analysis BrdU and -SMA were analyzed for the number of cells that stained positive. Each were viewed at 200x magnification and counted in 10 random fields. Days 15, 45 and 90 of AA and CCl4 treatment were utilized. Hepatocytes a nd non-hepatocytes were counted for BrdU. Activated hepatic stellate cells were counted for -SMA. Microsoft Excel wa s utilized to find the

PAGE 60

60 average and standard deviations and the pvalues were found using the t-test from http://www.graphpad.com. The significance between AA and CCl4 at days 15, 30, 45, 60, 70 and 90 for each hepatocyte and non-hepato cytes for BrdU and for stellate cells, days 15, 45 and 90 of treatment for -SMA were in put into the t-test. Table 2-1 List of antibodies utilized Primary Cat # Co. Dilution Retrieval Secondary BrdU M0744 Dako 1:200 None Donkey anti-mouse Fibronectin ab23751 Abcam 1:200 Tryp sin Digest Donkey anti-mouse CTGF ab6992 Abcam 1:800 None Goat anti-mouse Collagen IV ab13966 Abcam 1:100 Trypsin Digest Goat anti-mouse SMA A5228 Sigma 1:200 Trypsin Di gest Donkey anti-mouse Laminin Z0097 Dako 1:25 Proteinase K Goat anti-rabbit TGF sc-146 Santa Cruz 1:100 Citra Goat anti-rabbit AFP A0008 Dako 1:600 Trilogy Goat anti-rabbit CD31 sc-1506 Santa Cruz 1: 50 Citra Rabbit anti-goat vWF sc-8068 Santa Cruz 1: 600 Citra Rabbit anti-goat

PAGE 61

61 CHAPTER 3 RESULTS Optimizing Protein and Nucleic Acid Extracti ons from Archival Formalin Fixed Paraffin Embedded Tissue Despite repeated attempts at purification, electrophoretic resoluti on of discrete protein bands from archival tissue was not possible (Fig ure 3-1B). However, we attempted to resolve beta actin on Western blots of archival tissue. Fresh NRL di splayed a strong band of the appropriate molecular weight (F igure 3-1A). There was no corre sponding band in the archival tissue lane (Figure 3-1A). Intere stingly, fresh formalin fixed NRL also failed to produce a beta actin band by Western blot (Figure 3-1A). Amplification of a region of the GAPDH gene in DNA isolated from archival, paraffin embedded NRL was possible (Figure 3-1C). Howeve r, PCR amplification of DNA isolated from archival, paraffin embedded allyl alc ohol (AA) and carbon tetrachloride (CCl4) treated liver was less successful (Figure 3-1D). As double stranded DNA is much more stable than single stranded RNA, RT-PCR was not attempted on archival tissue. Assessing the Allyl Alcohol and Carbon Tetrachloride Chronic Liver Injury Models Tissue from formalin fixed, paraffin embedded blocks were cut into 5 m sections and stained. Time points from days 15, 30, 40, 45, 54, 60, 65, 68, 70 and 90 were used. Days 15, 30, 45, 60, 70 and 90 are shown in H&E. Immunohistoche mical data are presen ted from d15, 45 and 90. All chromagen stains were run with norma l rat liver (NRL) and a matching immunoglobulin controls. The samples were photographed using an Olympus microscope and Optronics digital camera Olympus, Melville, NY).

PAGE 62

62 Optimizing Protein and DNA Extractions from Archival Formalin Fixed Paraffin Embedded Tissue Despite repeated attempts at purification, electrophoretic resoluti on of discrete protein bands from archival tissue was not possible (Fig ure 3-1B). However, we attempted to resolve beta actin on Western blots of archival tissue. Fresh NRL di splayed a strong band of the appropriate molecular weight (F igure 3-1D). There was no corre sponding band in the archival fibrotic tissue lanes (Fig ure 3-1D). Interestingly, fresh form alin fixed NRL also failed to produce a beta actin band by Wester n blot (Figure 5A). Amplification of a region of the GAPDH gene in DNA isolated from archival, paraffin embedded NRL was possible (Figure 3-1C). Howeve r, PCR amplification of DNA isolated from archival, paraffin embedded AA and CCl4 treated liver was less succ essful (Figure 3-1D). As double stranded DNA is much more stable th an single stranded RNA, RT-PCR was not attempted on archival tissue. Overall Evaluation of the Two Injury Models As expected, the AA model showed signs of necrosis around the porta l triad (PT) with relatively little damage around the central vein (CV) (Figure 3-2 A6). In addition, ductular proliferation, connective tissu e accumulation and cirrhosis were evident in this model. Accumulation of small cells with morphology consis tent with oval cells was also seen in this model. In contrast to the AA model, CCl4 injury induced mainly cen trilobular necrosis. Also apparent in this model were duc tular proliferation, micro nodule formation, lipid storage within hepatocytes (steatosis) and massive fibr osis eventually leading to cirrhosis.

PAGE 63

63 Hepatocyte and Non-Parenchymal Cell Prol iferation in Chronic AA and CCl4 Models Cell proliferation during the time course of AA and CCl4 chronic injury was determined by BrdU incorporation into newly synthesized DNA. BrdU labeling of NRL and rat liver 24hr following PH were used as negative and positiv e controls, respectively. Analysis of the BrdU staining pattern in AA treated livers indicates co ncentrations of proliferating cells within localized pockets, highlighting th e need to count many different mi croscopic fields. This is best demonstrated in the non-parenchymal fraction of AA d90 (Figure 3-3) where several fields contained greater than 200 proliferating cel ls while the average was only 91. In the CCl4 model, two waves of hepatocyte prolif eration were noted with peaks at d45 and d90 (Table 3-1). The AA model also involved an ebb and flow with re spect to hepatocyte proliferation, though this was much less pronounced than in the CCl4 model. Non-parenchymal cell proliferation appears to lag behind hepatocyte prolifer ation in these models with a robust peak of proliferation noted at d90 in the AA model. As expected, normal rat liver shows very little prolif eration (Figure 3-4C). Partial hepatectomy rat liver, 24hr following surg ery demonstrates massive proliferation of the hepatocyte cell fraction (Figure 3-4D). A peak of non-parenchymal cell proliferation is noted several days later (data not shown). Alpha Fetoprotein Expression by Hepa tic Oval Cells in Chronic AA and CCl4 Models To determine the degree of oval cell participa tion in these two injury models, the oval slides were stained for the oval cell marker AFP. Adult NRL contains very few, if any, AFP+ cells, and was used as a negative control (Fi gure 3-5C). The AA model has previously been shown to induce an oval cell response. In agr eement with this noti on, a population of AFP+ hepatic oval cells develops over the time course of chronic AA injury (Figure 3-5A). Analysis of higher magnification fields shows the AFP+ oval cells in the AA model are arranges in duct like structures (Figure 3-5E). These st ructures are also found in the CCl4 model, though they are quite

PAGE 64

64 rare; oval cells are found scattere d throughout the fibrotic cord st ructures (Figure 3-5F). Also seen in the CCl4 model are several AFP+ transitional hepatocytes are ap parent at later time points (Figure 3-5B). Trasnsforming Growth Factor Beta Expressi on by Hepatic Cells in Chronic AA and CCl4 Models The profibrotic cytokine TGF was used to identify tissu e microenvironments conducive to the accumulation of ECM. Normal liver demonstrated very mild TGF expression that was more evident in the portal zone (Figure 3-6C). TGF was found to be expressed by a variety of cell types in each of the chr onic injury models. In the CCl4 model, TGF is strongly expressed by non-parenchymal cells that are in intimate cont act with fibrotic cords. A more diffuse TGF staining is evident in the populati on of small hepatocytes peripheral to the necrotic zone in early time points (Figure 3-6B1). Expression of TGF by these cells subsides by middle to late time points (Figure 3-6B2-3 and F). In the AA m odel, early time points indicate mild TGF expression by hepatocytes within the portal zone (Figure 3-6A1). In later time points, expression shifts to the non-parenchymal cell fraction within areas of fibrosis (Figure 3-6A2-3). Large areas of non-specific staining are apparent within the necrotic zones in later AA time points (Figure 36A3). Connective Tissue Growth Factor Expression by Hepatic Cells in Chronic AA and CCl4 Models CTGF has been shown to be a downstream mediator of TGF expression and was used as an additional indication of a pr o-fibrotic microenvironment (70) Early time points in the AA model shows punctate expression of CTGF by indi vidual hepatocytes within the central zone (Figure 3-7A1). Expression of this factor increases to involve a majority of the central zone hepatocytes in the mid time point s and eventually decreases by later time points (Figure 3-7A23). Also, occasional non-parenchymal cells within areas of fibrosis are positive for CTGF. This

PAGE 65

65 is especially true at the later time points (Figur e 3-7E). In contrast to the AA model, CTGF expression in the CCl4 model continually increases over the time course of injury (Figure 3-7 B1-3). Analysis of high magnification d90 slid es demonstrates CTGF expression both by hepatocytes as well as non pa renchymal cells (Figure 3-7F). NRL showed little staining concentrated mainly within the endothelial cells associated with the hepatic circulatory system (Figure 3-7C). Alpha-Smooth Muscle Actin Ex pression by Hepatic Stellate Cells in Chronic AA and CCl4 Models We used -SMA to identify activated HSCs in each chronic liver injury model. As expected, NRL strongly expresses -SMA in vessel walls and in occasional HSCs (Figure 38C1-2). Both the AA and CCl4 injury models de monstrate increasing -SMA expression by cells associated with areas of fibrosis throughout the time course of these st udies (Figure 3-8A1-3,B13). -SMA expression appears to be confined to within the fibr otic regions in the AA model while many -SMA+ cells are evident within the regene rating nodules found in later time points of the CCl4 model (Figure 3-8 E-F). Quantification of the -SMA+ cells within each of these models demonstrates a significantly greater number of HSCs in the CCl4 model as compared to the AA model (Figure 3-9 and Table 3-2). Collagen Expression by Hepati c Cells in Chronic AA and CCl4 Models Molecular signaling through TGF and CTGF promote collagen synthesis and deposition by HSCs. Interstitial collagen type s I, II and III (Figure 3-10) a nd type IV (Figure 3-11) were identified within the fibrotic areas in both the AA and CCl4 injury models. Trichrome staining demonstrated an initial increase in periportal, in terstitial collagen betw een the early and middle time points (Figure 3-10 A1-2). As the injury proc eeds, little additional collagen seems to be deposited (Figure 3-10 A2-3). Higher magnificati on images reveal light staining for collagen

PAGE 66

66 within the portal zone indicating a relatively diffuse ECM (Figure 3-10 D). IHC for collagen type IV shows a similar staining pattern in the AA injury model (Figure 3-11 A1-3). Interstitial collagen staining in the CCl4 model demonstrates progressive deposition of this ECM component throughout the time course of injury (Figure 3-10 B1-3). Staining appears to be much more intense in this model indicating a well organized, dense accumulation of ECM surrounding and bridging the central zones (Figure 3-10 E). The staining pattern for collagen IV is very similar to interstitial collagen with progres sive pericentral depos ition and concentration at the later time points (Figure 3-11 B1-3, F). Accumulations of collagen, clearly border the regenerative nodules in the CCl4 model, where as a very light deposition of collagen is present within the necrotic zones of the AA model (Figure 3-10 D, E). NRL displayed the expected staining patterns for these collagens with strong concentrations f ound bordering vessels and within the basement membrane of the biliary tree (Figures 3-10, 3-11 C1-2). Laminin Expression by Hepatic Cells in Chronic AA and CCl4 Models Laminin is another component of ECM. This ma trix protein has been shown to play a role in migration of cells to regions of injury. La minin staining was used as an indication of cell trafficking potential through fi brotic regions. NRL was used as a reference for comparison (Figure 3-12 C1-2). In the AA model, laminin doe s not appear to accumu late throughout the time course of injury (Figure 3-12 A1-3). Diffuse lami nin staining within the pe riportal necrotic zone is likely non-specific (Figure 3-12 A3, E). Conve rsely, strong progressive laminin deposition is apparent in the pericentral fibrotic regions of the CCl4 model (Figure 3-12 B1-3, F). Laminin has also been used to identify the organization of hepatic parenchyma. Ordered cords of hepatocytes are observed in NRL (Figure 3-12 C2). It is in teresting to note that hepatocytes within the regenerative nodules at later time points of the CCl4 injury do not appear to be well organized (Figure 3-12 F).

PAGE 67

67 Fibronectin Expression by Hepatic Cells in Chronic AA and CCl4 Models Fibronectin has been shown to be a major comp onent of the provisional matrix associated with oval cell proliferation (Pi, accepted fo r publication, Hepatology 2007). As there is a significant oval cell component to the regenerative response to AA injury as compared to CCl4 injury, fibronectin was stained in each of thes e models. Fibronectin deposition in the AA model becomes apparent by the mid time points. Dense tendr ils of fibronectin are seen to arborize from the portal zone into the hepatic parenchyma (Fi gure 3-13 A2). By later time points, fibronectin remains within the enlarged n ecrotic regions surroundi ng the portal veins, though organization is lost (Figure 3-13 A3, E). The CCl4 model demonstrates much less fibronectin deposition throughout the entire time course of injury (F igure 3-13 B1-3). The strongest indication of fibronectin may be seen within the basement membrane surrounding areas of proliferating bile ducts (Figure 3-13 B3, F). In NRL, fibronect in is found surrounding all vessels and ducts (Figures 3-13 C1 and 2). Week expression is no ted outside of these structures even by d15 in both models.

PAGE 68

68 Table 3-1 Proliferating (BrdU+) hepatocytes (H) and non-hepatocytes (NH) over the time course of chronic AA and CCl4 liver injuries. Column one of each H and NH represents the day of treatment, column two and three are the average number of cells positive for BrdU after AA and CCl4 treatment, and column four show s the statistical difference between the two chronic injury models represented by a p-value generate d from t-test analysis for each time point. H AA CCl4 P-Value NH AA CCl4 P-Value d15 7+/3 14+/-4 0.0003 d15 6+/-2 30+/-5 0.0001 d30 2+/-1 7+/-2 0.0001 d30 10+/-3 15+/-7 0.0525 d45 9+/-5 36+/-14 0.0001 d45 25+/-129+/-3 0.0007 d60 7+/-4 8+/-3 0.5350 d60 28+/-1027+/-6 0.7894 d70 6+/-10 30+/-7 0.0001 d70 31+/-1815+/-5 0.0144 d90 14+/-14 28+/-16 0.0518 d90 91+/-6120+/-4 0.0017 Table 3-2 Activation of HSCs over th e time course of chronic AA and CCl4. Column one shows the days of treatment. Column two represen ts the average number of HSCs in a 200x in field for AA. Column three represents the average number of HSCs in a 200x field for CCl4. Column four represents the p-value determined by a t-test between the two models during the time course. HSCs AA CCl4 P-value d15 1+/-2 27+/-8 0.0001 d45 27+/-7 74+/-20 0.0011 d90 59+/-11 82+/-17 0.0347

PAGE 69

69 Figure 3-1 Protein and Nucleic Acid analysis of AA and CCl4 archival paraffin embedded formalin fixed tissue. (A) Western blot analys is of beta actin in fresh normal rat liver tissue (FNRL), freshly paraffin embedded, form alin fixed normal rat liver (FNRLF), and archival paraffin embedded, formalin fixe d normal rat liver (ANRLF). (B) Coomassie blue staining was done on 3 different isolati on procedures of archival paraffin embedded, formalin fixed normal rat liver (NA1-3). NA1 represents protein with only basic purification. NA2 has two additional rounds of purification. NA3 has four additional rounds of purification. (C) PCR amplifica tion of GAPDH in DNA samples from fresh NRL (FT1-2), archival formalin fixed, paraffin embedded tissue (AP1-2) and fresh formalin fixed, paraffin embedded tissue (FP1 -2). (D) PCR amplification of GAPDH in purified DNA samples from fresh formalin fixed, paraffin embedded NRL (lane 2), archival AA treated rat liver (lane 3-9), ar chival formalin fixed, paraffin embedded NRL (lane 10), archival CCl4 treated rat liver (lane 11-14). C D B NA1 NA2 NA3 Ruler A APNRL FPNRL FNRL

PAGE 70

70 Figure 3-2 Hematoxylin and Eosi n analysis of AA (A) and CCl4 (B) treated rat livers 90 days following initiation of injury. Time cour se of chronic AA injury: A1=d15, A2=d30, A3=d45, A4=d60, A5=d70, A6=d90. Ti me course of chronic CCl4 injury: B1=d15, B2=d30, B3=d45, B4=d60, B5=d70, B6=d90. Labe ling points out the central vein (CV) and portal triad (PT). A and B, or iginal magnification= 200X. A1-B6, original magnification=100X. CV PT A CV PT B

PAGE 71

71 Figure 3-2 Continued A1 A4 A5 A2 A3 A6 B1 B4 B5 B2 B3 B6

PAGE 72

72 Figure 3-3 Proliferating (BrdU+) cells over the time course of chronic AA (blue) and CCl4 (pink) liver injuries.

PAGE 73

73 Figure 3-4 Immunohistochemical la beling of proliferating, BrdU+ nuclei. (A1-3) AA injury at d15, 45 and 90, respectively. (B1-3) CCl4 injury at d15, 45 and 90, respectively. (C) NRL (D) Rat liver, 24hr following partial hepatectomy. Original magnifications=200X. A1 B1 B2 A2 A3 B3 C D

PAGE 74

74 Figure 3-5 AFP+ hepatic oval cells in the portal regions of rat liver. (A1-3) AA injury at d15, 45 and 90, respectively. (B1-3) CCl4 injury at d15, 45 and 90, respectively. (C) NRL. (D) immunoglobulin control. (E-F) Highe r magnification of d90 AA and D90 CCl4, respectively. A and B original magnification=600X, C-F original magnification=1000X. E B1 B2 A2 A3 B3 F C D A 1

PAGE 75

75 Figure 3-6 TGF expression during chronic liver inju ry. (A1-3) AA injury at d15, 45 and 90, respectively. (B1-3) CCl4 injury at d15, 45 and 90, re spectively. (C1-2) NRL. (D) immunoglobulin control. (E-F) Higher magnification of d90 AA and d90 CCl4, respectively. A and B original magnificat ion=600X, C1 original magnification=200X, C2-D original magnification=600X, E-F original magnification=200X. E A1 B1 B2 A2 A3 B3 F C1 C2 D

PAGE 76

76 Figure 3-7 CTGF expression duri ng chronic liver injury. (A13) AA injury at d15, 45 and 90, respectively. (B1-3) CCl4 injury at d 15, 45 and 90, respectively. (C1-2) NRL. (D) immunoglobulin control. (E-F) Higher magnification of d90 AA and D90 CCl4, respectively. A and B original magnificat ion=600X, C1 original magnification=200X, C2-D original magnification=600X, E-F original magnification=200X. A1 B1 B2 A2 A3 B3 E F C1 C2 D

PAGE 77

77 Figure 3-8 Alpha-Smooth Muscle Actin ( -SMA) expression during chro nic liver injury. (A1-3) AA injury at d15, 45 and 90, respectivel y. (B1-3) CCl4 injury at d15, 45 and 90, respectively. (C1-2) NRL. (D) immunoglobu lin control. (E-F) Higher magnification of d90 AA and D90 CCl4, respectively. A a nd B original magnification=600X, C1 original magnification=200X C2-D original magnificat ion=600X, E-F original magnification=200X. E A1 B1 B2 A2 A3 B3 F C1 C2 D

PAGE 78

78 Figure 3-9 Activated HSCs over the time course of chronic AA (blue) and CCl4 (pink) liver injuries. Figure 3-10: Trichrome staining fo r Collagen I during ch ronic liver injury. (A1-3) AA injury at d15, 45 and 90, respect ively. (B1-3) CCl4 injury at d15, 45 and 90, respectively. (C1-2) NRL. (D-E) Higher magnification of d90 AA and D90 CCl4, respectively. A and B original magnificat ion=600X, C1 original magnification=200X, C2-F original magnification=600X.

PAGE 79

79 Figure 3-10 Trichrome staining for collagen I during chronic liver in jury. (A1-3) AA injury at d15, 45 and 90, respectively. (B1-3) CCl4 inju ry at d15, 45 and 90, respectively. (C1-2) NRL. (D-E) Higher magnification of d90 AA and D90 CCl4, respectively. A and B original magnification=600X, C1 origin al magnification=200X, C2-F original magnification=600X. A1 B1 B2 A2 A3 B3 D E C1 C2

PAGE 80

80 Figure 3-11 Collagen type IV in chronic liver injury models by immunohistochemistry. (A1-3) AA injury at d15, 45 and 90, respectivel y. (B1-3) CCl4 injury at d15, 45 and 90, respectively. (C1-2) NRL. (D) immunoglobu lin control. (E-F) Higher magnification of d90 AA and D90 CCl4, respectively. A a nd B original magnification=600X, C1 original magnification=200X, C2-F original magnification=600X. E F A1 B1 B2 A2 A3 B3 C1 C2 D

PAGE 81

81 Figure 3-12 Laminin deposition in chronic liver injury models by immunohistochemistry. (A13) AA injury at d15, 45 and 90, respectivel y. (B1-3) CCl4 injury at d15, 45 and 90, respectively. (C1-2) NRL. (D) immunoglobulin control. (E-F) Highe r magnification of d90 AA and D90 CCl4, respectively. A and B or iginal magnification=600X, C1 original magnification=200X, C2-F or iginal magnification=600X. A1 B1 B2 B3 A3 A2 E F C1 C2 D

PAGE 82

82 Figure 3-13 Fibronectin in chronic liver inju ry models by immunohistochemistry. (A1-3) AA injury at d15, 45 and 90, respectively. (B1-3) CCl4 injury at d15, 45 and 90, respectively. (C1-2) NRL. (D) immunoglobu lin control. (E-F) Higher magnification of d90 AA and D90 CCl4, respectively. A-C1 original magnification=200X, C2-F original magnification=400X. F E C1 C2 D A1 B1 B2 A2 A3 B3

PAGE 83

83 CHAPTER 4 DISCUSSION The overall goal of this project was to invest igate differences in th e fibrotic response to chronic hepatic portal zone injury versus chroni c central zone injury. Deposition of ECM in each of these injuries is orchestrated by an extensive series of molecular signals, which interact with inflammatory cells as well as fibroblasts. Clas sical wound repair theory dictates that the inflammatory cell component of the response to injury governs the activit y of tissue fibroblasts which, in turn, synthesize ECM as a scaffold on which new cells may migrate and proliferate. In the context of chronic injury, this process often goes awry resulting in the inappropriate accumulation of acellular matrix or a fibrotic "scar". Our models of chronic liver injury demonstrate that this process may be significantly different de pending on the location of injury across the hepatic lobule. Analysis of Archival Tissue Initially, we attempted to purify protein suitab le for Western blot analysis from archival formalin fixed, paraffin embedded tissue. This would have benefited the current study by allowing for quantification of molecular signals and ECM components from the vast archive of injury models accumulated in this laboratory. The initial challenge was to extract wax free tissue from paraffin blocks. This proved to be easily accom plished by heating sections of tissue (up to 40 m) in xylene for up to 1 hour. The fixed tissue pr oved to be fairly resilient to SDS treatment and a solution of 2% SDS was required to comp letely solubilize tissue homogenates. We found that supernatants from tissues processed in this manner contained protein that could be precipitated with cold acetone. We soon discovered that the proteins isolated from these samples were not usable for Western blot analysis (Fi gure 3-1 A). It is likely that residual crosslinking of protein, a result of

PAGE 84

84 the initial formalin fixation, was hampering migra tion through the gel. Acid ification or alkylation is a well known method for disrupting fo rmalin crosslinking of proteins for immunohistochemistry. Samples were treated with bu ffers of various pH w ith or without a cation chelators. However, proteins still could not be reproducibly resolved by electrophoresis (Figure 3-1 B). Attempts to isolate amplifiable DNA from ar chival tissue met with marginal success (Figure 3-1 C, D). Faint bands of amplified DNA were identifiable follow ing PCR amplification. The isolation procedure proved to be unreliable in terms of reproducibility. RT-PCR of RNA from archival tissue seemed very unlikely, so we chose to concentrate on immunohistochemistry to describe the fibrotic res ponse to chronic liver injury. Immunohistochemical Analys is of Archival Tissue Analysis of H&E stained tissues gave the expected results (Figure 3-2). Chronic AA induced periportal necrosis and CCl4 induced pericentral necrosis. We were intrigued that early time points form the AA treated animals demonstrated a very mild initial injury as compared to the CCl4 model. This was somewhat surprising as these animals appeared to be much sicker throughout the course of treatment. Later time point s provided an answer to this contradiction. While the AA injury induced a minor initial in jury, damage continued to accumulate over the course of treatment. In contrast to this, the CCl4 model induced relatively severe injury in the early to mid time points (Figure 3-2). However, as treatment continued, the rate of tissue destruction appeared to slow. A possible explanation for this lies in the toxicology underlying each of these injury models. AA is metabolized by the portal zone cells wh ich may be considered relatively immature and have a low ploidy of only 2N. Oval cells that proliferate from the portal zone subsequently differentiate into immature hepatocytes that are able to metabolize AA. These cells would be

PAGE 85

85 subjected to the toxic metabolite and die. Additi onally, mature hepatocytes that enter the cell cycle have been shown to transiently adopt an immature phenotype with respect to xenobiotic metabolism [167]. This may render hepatocytes in the mid and cen tral zones susceptible to AA induced damage. The ploidy of the cell determines how quickly the cell can replicate; the fewer the copies of chromosomes, the more quickly th e cell can divide. Since the periportal cells are damaged, they are not able to repopulate the zone and the mid and pericentral zone cells are left to repopulate the liver. The closest cells are in the mid zone and have an average ploidy of 16N which means the cells would have to copy 8 times as many chromosomes as the original periportal cells. It is however possible to divide without replication and yield two 8N cells, but this is a less frequent event. The fact that these cells take longer to divide means that a lag time may occur, and by the time the cells replicate, another dose of AA is given which damages more cells and the liver may not able to heal. There may be another reason why we did not se e an attempt to heal the liver. A study by Jung in 2000 reported that a certain dose was ne eded to yield the pr oper response (36mg/kg body weight, given twice weekly for up to 16 weeks) [94]. Jung used Sprauge-Dawely rats which are known to show different responses to chemical s than the Fischer-344 rats. Our study utilized 7.4mg/kg body weight given 5 days consecutively with 2 days consecutively off for up to 13 weeks. Perhaps their procedur es lead to the development of reactive oxygen species which stimulate the hepatic stellate cells to lay down a pr otective scar to prevent further exposure of the chemical to the enclosed area. The most stri king aspect of the AA model is the lack of proliferating hepatocytes (Figure 3-3,4). Th is phenomenon is difficult to explain, and was evident in each AA treated animal examined throughout this study. Even though it would take longer to divide, these hepatocytes should still be proliferating. As mentioned above, it is

PAGE 86

86 possible that as mature hepatocytes enter the cell cycle, they become transiently able to metabolize AA and are either destroyed or their proliferative ability is diminished. Since the periportal tissue is damaged, oxygen and nutrients are not able to reach the re st of the liver by the later time points leading to addi tional stress and necrosis. This ma ssive necrosis that develops in these animals over time recruits inflammatory cells. There is robus t proliferation of nonparenchymal cells in the AA model, particularly at the later time points. We suspect that these inflammatory cells are responsible for the high ra te of proliferation seen in the non-parenchymal fraction. In contrast to our AA ch ronic injury model, CCl4 is metabolized by mature central zone hepatocytes with an average of 4N Destruction of these cells woul d result in the proliferation of oval cells and portal zone hepatocy tes that would, for a time, be unable to metabolize the toxin. The periportal cells, as mentions above are on average only 2N and can rapidly divide to repopulate the liver. Proliferation data shows that the CCl4 model induces robust proliferation of hepatocytes by d45 followed by a lack of prolif erating cells by d60 then a new surge of hepatocyte proliferation is seen by d90 within the rege nerating nodules (Figure 3-3). This is the expected regenerative response to the loss of hepatic tissue. There is modest proliferation of the non-parenchymal cell fraction in th is model through all time point s. These cells likely contain inflammatory cells recruited to the necrotic areas as well as fibroblasts and oval cells that are participating in tissue regeneration and rem odeling. Because the periportal tissue is not completely damaged, CCl4 injured liver can still receive nutri ents and oxygen, except perhaps to the hepatic foci which were created by the res ponse to the reactive oxygen species created from the metabolism of this chemical as seen by J ung for AA and by the numerous studies of chronic CCl4 treatment. These walled off foci might be ca lled regenerating nodules in this case because,

PAGE 87

87 at later time points in this treatment, activated hepatic stellate cells a nd ECM proteins are seen infiltrating the atypical hepatocytes bundle. An overall evaluation of the oval cell response in these two injury models shows a greater number of AFP+ cells in chronic AA (Figure 3-5). This is consistent with published research demonstrating that AA is able to recruit liver progenitor cells to the portal zone to affect repair. Interpretation of this data is complicated by severa l facts. The first of thes e is that only a portion of the oval cell population is AFP+. We can not rule out the possibility that the CCl4 model induces a population of oval cells that do not e xpress AFP. Additional markers for oval cells are known, but antibodies for these markers do not work on formalin fixed tissue. The second complicating factor involves the morphology of the AFP+ cells seen in later time points of AA injury. These cells do not display the classical ov al nuclei. Histologically, these cells resemble mature cholangiocytes. Because these cells are ge nerally arranged into ducts that are within a fibrous matrix, and do not make contact with hepa tocytes, we feel that these cells most likely represent early progeny of oval cells. They likely have moved toward the cholangiocyte lineage, but retain expression of AFP. On e note in this stain is that oval cells are a heterogeneous population in which not all oval cells express AF P, perhaps these are further differentiated. Despite the problems with oval cell quantification a ssociated with using AFP as a marker, it may be safely concluded that each of these models induces a degree of oval cell proliferation. The molecular microenvironment in regions of injury plays an important role in mediating the repair of damaged tissue. While not a direct part of this study, inflammatory cells are known to be the ultimate source of a ma jority of these molecular signals. TGF is a known regulator of hepatic stellate ce ll phenotype. As such, we analyzed slides from each injury model for TGF expression by immunohistochemist ry. The expression pattern for this cytokine differed

PAGE 88

88 greatly between these two models. In the CCl4 model, pockets of non-parenchymal cells within the zones of injury strongly express TGF (Figure 3-6 B1-3). This is li kely the result of cytokine expression by inflammatory cells recruited to thes e areas. Less intense staining is evident within a sub-population of hepatocytes within the unda maged parenchyma. Numerous pycnotic nuclei are present in this hepatocyte population bringing into question the specificity of this staining (Figure 3-6 B1); it has been noted that hepatocytes may take up latent TGF from the circulation and the large amount seen here (Figure 3-6 A1) can lead to apoptosis [136]. As the injury progresses, the necrotic central zone be gins to contract, c oncentrating the TGF + nonparenchymal cells into the developing fibrotic area. TGF + hepatocytes are seen infrequently at these later time points. Early time points in the AA injury model bear a striking similarity to the CCl4 model (Figure 3-6 A1, B1). Howe ver, mid to late time points TGF staining remains relatively diffuse within poorly or ganized zones that resemble gra nulation tissue (F igure 3-6 A3). There also appeared to be a slight decrease in total TGF in the later time points of the AA model. As expected, th e distribution of TGF expressing cells in each of these models correlates with the regions of developing fibrosis. CTGF has been shown to be a downstream mediator of TGF with respect to hepatic stellate cell activation [163]. We expected to see CTGF expresse d in a similar pattern as TGF The AA model showed occasional expression of CTGF by mid zone hepatocytes (Figure 3-7 A2). The number of hepatocytes expressing CTGF in creased to involve a majority of the healthy hepatocytes by the middle time points. Expression of this factor diminished over time, with only sporadic expression seen at the final time point. Interestingly, this decrease in expression at the later time points correlated with the decrease in TGF expression noted above. CTGF expression in the CCl4 model was also found predominantly with in the hepatocyte population. As with

PAGE 89

89 TGF CTGF expression was seen to consistently increase throughout th e time course of the study. While CTGF has classically been considered to be produced mainly by hepatic stellate cells, recent studies confirm that hepatocytes may express this factor in response to hepatic injury [88]. In our inju ry models, hepatocytes do, indeed, appe ar to be the major source of CTGF (Figure 3-7). TGF and CTGF are critical in the activation of hepatic stellate cells to a matrix synthesizing phenotype. A hallmark of hepatic st ellate cell activation is the expression of smooth muscle actin (SMA). In the CCl4 model, -SMA+ cells are easily identified, even in the earliest time points (Figure 3-8 B1-3). The number of these cells generally increases over the course of injury. -SMA+ cells are seen throughout the fibrotic regions of the liver particularly within the fibrous cords that encircle the regenerative nodule s. It is worth noting that -SMA+ hepatic stellate cells are found inf iltrating these nodules in the later time points (Figure 3-8 F). It is possible that these stellate cells provide matrix for organization of the hepatic microarchitecture within the nodul es. In contrast to the CCl4 model, the AA model shows very few SMA+ stellate cells in the early time points. While the number of these cells does increase over the time course of injury, it never reaches the numbers seen in the CCl4 model (Figure 3-8 A1-3, Figure 3-9). The stellate cells ar e the primary source of ECM seen in these models; it is not surprising that fewer stellate cells are present in the less fibrotic AA model. It is also possible that the -SMA+ fibroblasts in the AA model are not of st ellate cell origin but are myofibroblasts that surround bile ducts and vesse ls in the portal triad [168] and may not perform in the same manner. Hepatic stellate cell activation ultimatel y leads to matrix deposition. The major constituents of this matrix are collagens, lami nins and fibronectin. Th e pathologic accumulation

PAGE 90

90 of these proteins results in a fi brotic scar that diminishes live r function. When this progresses to cirrhosis, the life of the organism is jeopardized. Each of the aforementioned matrix proteins were assessed in our chronic injury models. Co llagen represents the most abundant protein found in the ECM. In both the AA and CCl4 models, collagen progressively accumulates at the zone of injury (Figures 3-10,11). Trichr ome staining shows that the colla gen that accumulates in the periportal zone in the AA model is much less densely arranged than in the CCl4 model (Figure 310 A1-3). This is an interesting observation as the AA model shows very little accumulation of collagen IV, which is known to crosslink colla gen I into tighter bundles (Figure 3-11 A1-3). Laminin is also much less abunda nt in the AA model than the CCl4 model. Laminin has been shown to bind to collagen IV, so it is not surprising that very little laminin is present in the injured zones of the AA model which are also de void of collagen IV (Fig ure 3-12 A1-3). In the mid time points, many regions of fibronectin de position may be seen radiating from the portal zone into healthy hepatic parenchyma (Figure 3-13 A2). At these same time points, many AFP+ oval cells are present in the portal zone. Our labo ratory has previously shown that a fibronectin rich provisional matrix is associated with prolif erating oval cells, and we suspect this is driving the deposition of fibronec tin in the AA model. Overall, it appears that collagen I is produced by stellate cell s at the zone of injury in each of the models included in this study. Incr eased profibrotic cytokine activity in the pericentral, CCl4 injury model correlates with robust stel late cell activation within the damaged zone. These activated fibroblasts secrete a dense matrix of collagen I th at is cross-linked by collagen IV which is bound with laminin. However, matrix modification by collagen IV occurs only in the CCl4 model resulting in a much tighter matr ix as compared to the AA model. The dense nature of the fibrotic scar may preven t remodeling by proteinase s allowing this ECM to

PAGE 91

91 persist. As with cirrhosis in the human liver, this matrix decreases li ver function, forcing an attempt at compensation which is evidenced by the development of rege nerative nodules within the matrix. However, these nodules lack appropria te organization and do little to resolve the situation. Matrix accumulation in the AA m odel is not as extensive as in the CCl4 model. Despite this fact, chronic AA administration results in a much sicker animal than chronic CCl4. This is likely due to continued injury by AA in the portal zone which, over time, enlarges the necrotic area. The CCl4 induced injury does not appear to wors en significantly at later time points suggesting that the remaining hepatocytes ma y not metabolize the toxin. From a clinical standpoint, this would suggest th at chronic portal zone injury may be more easily managed so long as the source of injury is identified and removed. Central z one injury may be less damaging to the hepatic parenchyma over time, though the resu lting fibrosis is persis tent and may continue to impair liver function even after the source of injury is removed.

PAGE 92

92 CHAPTER 5 FUTURE DIRECTIONS This study would benefit from utilizing in situ hybridization for TGF CTGF, Col I and IV, FN and LM. This would show where and which cells are produci ng these cytokines and proteins. In addition to these alr eady studied proteins, integrins th at bind cells to the ECM can be stained for. Also, since immune cells are so impo rtant to this model, KC neutrophil and T cellspecific markers would also be utilized. In ad dition, MMPs and TIMPs need to be studied through IHC and in situ hybridization. It is anticipated that by understanding the molecular mechanisms responsible for stellate cell proliferation and ex cess ECM production new therapeutic targets will be identified for the treatment of liver fibrosis. The obvious future direction would be to repeat this experiment to acquire fresh tissue to quantify RNA and protein expressi on. Also, treatment would be considered and more animals would be added to alleviate fibros is and necrosis at varying leve ls of disease. Treatment could include the addition of MMPs, siRNA, cell tran splantation (oval cells and other bone marrow cells). The use of protei nases would seem to be the most l ogical first step to overcome these diseases. If this is not sufficient, interruption of the ECM proteins at the translational step would be next. If this fails, cell tran splantation therapies would be c onsidered. The capacity of bone marrow cells to differentiate into hepatocytes and intestinal cells was confirmed in male to female human transplants. These cells are easier to obtain than ot her tissue-specific stem cells, and more have more plasticity. This plasticity co uld be a problem and since oval cells in the bone marrow could provide a more specific liver progenito r cell. Oval cells are still rather plastic, so hepatocytes would also be tested.

PAGE 93

93 LIST OF REFERENCES 1. Hepatology: A Textbook of Liver Disease. B oyer and Zakim (eds), Saunders. New York (2003). 2. Crawford JM, Robbins and Cotran. Pathologic Basis of Disease. Kumar,V., Abbas,A. & Fausto N. (eds.), pp. 877-938, Elsevi er Saunders, Philadelphia (2005). 3. Rappaport AM. Acinar units and the pat hophysiology of the liver. In: The liver morphology, biochemistry, physiology. Vol 1. Rouiller CH. (eds.), pp. 265-328, Academic Press, New York. (1963). 4. Saxena R, Theise ND, Crawford JM. Micr oanatomy of the human liver-exploring the hidden interfaces. Hepatology. 1999;30(6):1339-46. 5. Fausto N. The Liver biology and pathobiology. Arias,I.M. and Boyer,J.L. (eds.), pp. 1501-1518, Raven Press, New York (1994). 6. Overturf K, al-Dhalimy M, Ou CN, Finegol d M, Grompe M. Serial transplantation reveals the stem-cell-like regene rative potential of adult mouse hepatocytes. Am J Pathol. 1997;151(5):1273-80. 7. Tateno C, Yoshizato K. Growth potential and differentiation capacity of adult rat hepatocytes in vitro. Wound Re pair Regen. 1999;7(1):36-44. 8. Ogi M, Yokomori, Oda M, Yoshimura K, Nomura M, Ohshima S, Akita M, Toda K, Ishii H. Distribution and localization of caveolin -1 in sinusoidal cells in rat liver. Med Electron Microsc. 2003;36(1):33-40. 9. Kurose I. Miura S, Saito H, et al. Rat Kupffer cell-derived nitric oxide modulates induction of lymphokine-activated killer cell. Gastroenterology 1995;109:1958-1968. 10. Paul SC, Lv P, Xiao YJ, An P, Liu SQ, Luo HS. Thalidomide in rat liver cirrhosis: blockade of tumor necrosis f actor-alpha via inhibition of de gradation of an inhibitor of nuclear factor-kappaB. Pa thobiology. 2006;73(2):82-92. 11. Edwards MJ, Keller BJ, Kauffman FC, and Thurman RG. The involvement of Kupffer cells in carbon tetrachloride toxicit y. Toxicol Appl Pharmacol 1993;119:275-279. 12. Winau F, Hegasy G, Weiskirchen R, Weber S, Cassan C, Sieling PA, Modlin RL, Liblau RS, Gressner AM, Kaufmann SH. Ito cells are liver-resident antigen-presenting cells for activating T cell responses. Immunity 2007;26(1):117-29. 13. Roth S, Michel K, Gressner AM. (Latent) transforming growth factor-beta in liver parenchymal cells, its injury-dependent releas e and paracrine effect s on hepatic stellate cells. Hepatology 1998;27:1003.

PAGE 94

94 14. Kossmann T, Manthey CL, Brandes ME, Morganti-Kossmann MC, Ohura K, Allen JB, Mergenhagen SE, Wahl SM. Kupffer cells expr ess type I TGF-beta receptors, migrate to TGF-beta and participate in streptococcal ce ll wall induced hepatic granuloma formation. Growth Factors 1992;7(1):73-83. 15. Iredale JR, Benyon RC, Arthur MJ, Ferris WE, Alcolado R, Winwood PJ, et al. Tissue inhibitor of metalloproteinase-I messenger RNA expression is enha nced relative to interstitial collagenase messenger RNA in experimental liver injury and fibrosis. Hepatology 1996;24:176-84. 16. Knittel T, Mehde M, Kobold D, Saile B, Dint er C, Ramadori G. Expression patterns of matrix metalloproteinases and their inhibito rs in parenchymal and non-parenchymal cells of rat liver: regulation by TNF-alpha and TGF-beta1. J Hepatol. 1999;30(1):48-60. 17. Brandes ME, Mai UE, Ohura K, Wahl SM. Type I transforming growth factor-beta receptors on neutrophils mediate chemotaxis to transforming grow th factor-beta. J Immunol. 1991;147(5):1600-6. 18. Grotendorst GR, Smale G, Pencev D. Productio n of transforming growth factor beta by human peripheral blood monocyt es and neutrophils. J Cell Physiol. 1989;140(2):396-402. 19. Liver Growth and Repair. Strain A, Dieh l AM (eds.) New York. Chapman and Hall (1998). 20. Neubauer K, Knittel T, Aurisch S, Fellmer P, Ramadori G. Glial fibri llary acidic protein-a cell type specific marker for Ito cells in vi vo and in vitro. J Hepato l. 1996;24(6):719-30. 21. Geerts A. History, heterogeneity, devel opmental biology, and functions of quiescent hepatic stellate cells. Semin Liver Dis. 2001;21(3):311-35. 22. Wahl SM, Allen JB, Weeks BS, Wong HL, Kl otman PE. Transforming growth factor beta enhances integrin e xpression and type IV collage nase secretion in human monocytes. Proc. Na. Ac. Sci. USA 1993;90:4577-458. 23. Balazovich KJ, Fernandez R, Hinkovska -Galcheva V, Suchard SJ, Boxer LA. Transforming growth factor-beta1 stimula tes degranulation a nd oxidant release by adherent human neutrophils. J Leukoc Biol. 1996;60(6):772-7. 24. Arthur MJP. Fibrogenesis II. Metalloproteinases and their inhibitors in liver fibrosis. Am J Physiol Gastrointest Liver Physiol 2000;279:245-249. 25. Wheeler MD, et al. The role of Kupffer ce ll oxidant production in early ethanol-induced liver disease. Free Radical Bi ology & Medicine 2001;31:12:1544-1549.

PAGE 95

95 26. Saperstein LA, Jirtle RL, Farouk M, Thompson H J, Chung KS, Meyers W C. Transforming growth factor-b 1 and mannos e 6-phosphate/insulin-lik e growth factor-II receptor expression during intrahepatic bile duct hyperplasia and biliary fibrosis in the rat. Hepatology 1994;19:412-417. 27. Sedlaczek N, Jia JD, Bauer M, Herbst H, Ruehl M, Hahn EG, Schuppan D. Proliferating bile duct epithelial cells are a major source of connective tissue gr owth factor in rat biliary fibrosis. Am J Pa thol. 2001;158(4):1239-44. 28. Michalopoulos GK, Barua L, Bowen WC. Transd ifferentiation of ra t hepatocytes into biliary cells after bi le duct ligation and toxic bili ary injury. Hepatology 2005;41(3):53544. 29. Farber E. Similarities in the sequence of early histologic changes induc ed in the liver of the rate by ethionine, 2AAF, and 3-methyl-4-dimethylaminoazbenzene. Cancer Res. 1956;16:142. 30. Michalopoulos GK, DeFrances MC. Liver re generation. Science. 1997;276(5309):60-6. 31. Oh SH, Muzzonigro TM, Bae SH, LaPlante JM, Hatch HM, Petersen BE. Adult bone marrow-derived cells trans-differentiating into insulin-producing cells for the treatment of type I diabetes. Lab Invest. 2004;84(5):607-17. 32. Hixson DC, Allison JP. Monoclonal antibodies r ecognizing oval cells induced in the liver of rats by N-2-fluorenylacetam ide or ethionine in a cholin e-deficient diet. Cancer Res 1985;45:3750. 33. Petersen BE, Zajac FV, Michalopoulos GK. Hepa tic oval cell activa tion in response to injury following chemically induced periporta l or pericentral damage in rats. Hepatology 1998;27:1030. 34. Oh SH, Hatch HM, Petersen BE. Hepatic oval stem cell in liver regeneration. Seminars in CDB. 2002;13:405-409. 35. Petersen BE, Zajac VF, Michalopoulos GK. Ex tracellular matrix rem odeling at the early stages of liver regeneration in the rat. Am J Pathol. 1997;151:905. 36. Shinozuka H, Lombardi B, Sell S, Iamma rino RM. Early histological and functional alterations of ethionine liver carcinogenesis in rats fed a choline-deficient diet. Cancer Res. 1978;38(4):1092-8. 37. Hixson DC, Chapman L, McBrid e A, Faris R, Yang L. Antigenic phenotypes common to rat oval cells, primary hepatocellular carcinomas and developing bile ducts. Carcinogenesis. 1997;18(6):1169-75.

PAGE 96

96 38. Nagy P, Kiss A, Schnur J, Thorgeirsson SS. De xamethasone inhibits the proliferation of hepatocytes and oval cells but not bile duct cells in rat liver. He patology 1998;28(2):4239. 39. Omori M, Evarts RP, Omori N, Hu Z, Ma rsden ER, Thorgeirsson SS. Expression of alpha-fetoprotein and stem ce ll factor/c-kit system in b ile duct ligated young rats. Hepatology 1997;25(5):1115-22. 40. Omori N, Omori M, Evarts RP, Teramoto T, Miller MJ, Hoang TN, Thorgeirsson SS. Partial cloning of rat CD34 cDNA and expres sion during stem cell-dependent liver regeneration in the adult ra t. Hepatology 1997;26(3):720-7. 41. Fujio K, Evarts RP, Hu Z, Marsden ER, Thor geirsson SS. Expression of stem cell factor and its receptor, c-kit, during liver regenerati on from putative stem cells in adult rat. Lab Invest. 1994;70(4):511-6. 42. Golding M, Sarraf CE, Lalani EN, Anilkumar TV, Edwards RJ, Nagy P, Thorgeirsson SS, Alison MR. Oval cell differentiation into hepatocytes in the acetylamino-fluorene (2AAF) treated regenerating rat liver. Hepatology 1995;22:1243. 43. Evarts RP, Nagy P, Nakatsukas H, Marsden E, Thorgeirsson SS. In vivo differentiation of rat liver oval cells into hepatocy tes. Cancer Res. 1989;49:1541. 44. Fausto N, Campbell JS. Mech The role of hepa tocytes and oval cells in liver regeneration and repopulation. Dev. 2003;120(1):117-30. 45. Nagy P, Bisgaard HC, Santoni-Rugiu E, Thorge irsson SS. In vivo infusion of growth factors enhances the mitogenic response of rat hepatic ductal (oval) cells after administration of 2-acetylaminofluorene. Hepatology 1996;23(1):71-9. 46. Hasuike et al Hepatocyte growth factor accelerates the proliferation of hepatic oval cells and possibly promotes the differentiation in a 2-acetylaminofluorene/partial hepatectomy model in rats. Journal of Gastroenterology and Hepatology 2005;20:1753. 47. Petersen BE, et al Bone Marrow as a Potential Sour ce of Hepatic Oval Cells. Science 1999;284. 48. Martinez-Hernandez A, Amenta PS. The extr acellular matrix in hepatic regeneration. FASEB.1995;9. 49. Higgins GM and Anderson RM. Experimental pa thology of the liver:I. Restoration of the liver of the white rat following partial su rgical removal. Arch Pathol. 1931;12:186-202.

PAGE 97

97 50. Lindroos PM, Zarnegar R, Michalopoulos GK. Hepatocyte growth factor (hepatopoietin A) rapidly increases in plas ma before DNA synthesis and liv er regeneration stimulated by partial hepatectomy and carbon tetrachlorid e administration. He patology 1991;13(4):74350. 51. Jirtle RL, Meyer SA. Liver tumor promotion: effect of phenobarbit al on EGF and protein kinase C signal transduction and transforming growth fact or-beta 1 expression. Dig Dis Sci. 1991;36(5):659-68. 52. Wu H, Wade M, Krall L, Grisham J, Xiong Y, Van Dyke T. Target ed in vivo expression of the cyclin-dependent kinase inhibito r p21 halts hepatocyte cell-cycle progression, postnatal liver development and rege neration. Genes Dev. 1996;10(3):245-60. 53. Mead JE, Fausto N. Transforming growth f actor alpha may be a physiological regulator of liver regeneration by means of an autocr ine mechanism. Proc Natl Acad Sci USA. 1989;86(5):1558-62. 54. Cressman DE, Greenbaum LE, DeAngelis RA, Ciliberto G, Furth EE, Poli V, Taub R. Liver failure and defective hepatocyte rege neration in interleukin-6-deficient mice. Science. 1996;274(5291):1379-83. 55. Yamada Y, Kirillova I, Peschon JJ, Fausto N. Initiation of liver growth by tumor necrosis factor: deficient liver regeneration in mice lack ing type I tumor necrosis factor receptor. Proc Natl Acad Sci USA. 1997;94(4):1441-6. 56. Yamada Y, Fausto N. Deficient liver regeneration after carbon tetrachloride injury in mice lacking type 1 but not type 2 tumor necrosis factor receptor. Am J Pathol. 1998;152(6):1577-89. 57. Alison MR, Golding MH, Sarraf CE. Pluripotential liver stem cells: facultative stem cells located in the biliary tree Cell Prolif. 1996;29(7):373-402. 58. Sirica AE, Mathis GA, Sano N, Elmore LW. Isolation, culture, and transplantation of intrahepatic biliary epith elial cells and ova l cells. Pathobiology 1990;58(1):44-64. 59. Thorgeirsson SS. Hepatic stem cells in liver regeneration. FASEB J. 1996;10:1249-1256. 60. Paku S, Schnur J, Nagy P, Thorgeirsson SS. Orig in and structural evolution of the early proliferating oval cells in rat liver. Am J Pa thol. 2001;158(4):1313-23. 61. Schuppan D, Ruehl M, Somasundaram R, Hahn E. Matrix as a modulator of hepatic fibrogenesis. Seminars in Liver Disease. 2001;21:3. 62. Lin CQ, Bissell DM. Multi-faceted regulation of cell differentiation by extracellular matrix. FASEB.1993;7.

PAGE 98

98 63. Paralkar VM, Vukicevic S, Reddi AH. Transfor ming growth factor beta type 1 binds to collagen IV of basement membrane matrix: implications for development. Dev Biol. 1991;143(2):303-8. 64. Zarnegar R, Michalopoulos GK. The many faces of hepatocyte growth factor: from hepatopoiesis to hematopoiesis. J Cell Biol. 1995;129(5):1177-80. 65. Yang C, Zeisberg M, Mosterman B, Sudhakar A, Yerramalla U, Holthaus K, Xu L, Eng F, Afdhal N, Kalluri R. Liver fibrosis: insights into migration of hepa tic stellate cells in response to extracellular matrix and growth factors. Gastroenterology 2003;124(1):14759. 66. Skalli O, Pelte MF, Peclet MC, Gabbiani G, G ugliotta P, Bussolati G, Ravazzola M, Orci L. Alpha-smooth muscle actin, a differentiation marker of smooth muscle cells, is present in microfilamentous bundles of pericytes. J Histochem Cytochem. 1989;37(3):315-21. 67. Hautmann MB, Madsen CS, Owens GK. A tran sforming growth factor beta (TGFbeta) control element drives TGFbeta-induced stim ulation of smooth muscle alpha-actin gene expression in concert with two CArG el ements. J Biol Chem. 1997;272(16):10948-56. 68. Roberts AB, Sporn MB, Assoian RK, Smith JM, Roche NS, Wakefield LM, Heine UI, Liotta LA, Falanga V, Kehrl JH, et al. Tr ansforming growth factor type beta: rapid induction of fibrosis and angiogenesis in vi vo and stimulation of collagen formation in vitro. Proc Natl Acad Sci U S A. 1986;83(12):4167-71. 69. Bornstein P, Sage H. Regulation of collage n gene expression. Prog Nucleic Acid Res Mol Biol. 1989;37:67-106. 70. Pi L, Oh SH, Shupe T, Petersen BE. Role of connective tissue growth factor in oval cell response during liver regeneration after 2-AAF/PHx in rats. Gastroenterology 2005;128(7):2077-88. 71. Gui Y, Murphy LJ. Insulin-like growth factor (IGF)-binding protei n-3 (IGFBP-3) binds to fibronectin (FN): demonstration of IGF -I/IGFBP-3/fn ternary complexes in human plasma. J Clin Endocrinol Metab. 2001;86(5):2104-10. 72. Wewer UM, Engvall E, Paulsson M, Yamada Y, Albrechsten R. Laminin A, B1, B2, S and M subunits in the postnatal rat liver de velopment and after partial hepatectomy. Lab Invest. 1992;66(3):378-89. 73. Fox JW, Mayer U, Nischt R, Aumailley M, Reinhardt D, Wiedemann H, Mann K, Timpl R, Krieg T, Engel J, et al. Recombinant ni dogen consists of three globular domains and mediates binding of laminin to collage n type IV. EMBO J. 1991;10(11):3137-46. 74. McKee KK, Harrison D, Capizzi S, Yurchenc o PD. Role of laminin terminal globular domains in basement membrane asse mbly. J Biol Chem. 2007;282(29):21437-47.

PAGE 99

99 75. Shakado S, Sakisaka S, Noguchi K, Yoshita ke M, Harada M, Mimura Y, Sata M, Tanikawa K. Effects of extracellular matri ces on tube formation of cultured rat hepatic sinusoidal endothelial cells Hepatology 1995;22(3):969-73. 76. Kato S, Otsu K, Ohtake K, Kimura Y, Yashiro T, Suzuki T, Akamatsu N. Concurrent changes in sinusoidal expression of laminin a nd affinity of hepatocytes to laminin during rat liver regeneration. Exp Cell Res. 1992;198(1):59-68. 77. Carlsson R, Engvall E, Freeman A, Ruosla hti E. Laminin and fibronectin in cell adhesion: enhanced adhesion of cells from regenerating liver to laminin. Proc Natl Acad Sci USA. 1981;78(4):2403-6. 78. Gressner AM, Weiskirchen R. Modern pathogen etic concepts of liver fibrosis suggest stellate cells and TGFas major players and therapeutic targets. J. Cell. Mol. Med. 2006;10:1:76-99. 79. Bataller R, Brenner DA. Liver fibrosis. The Journal of Clin ical Investigation 2005;115:2:209-218. 80. Anthony PP, Ishak KG, Nayak NC, Poulsen HE Scheuer PJ, Sobin LH. The morphology of cirrhosis. Recommendations on definiti on, nomenclature, and classification by a working group sponsored by the World Heal th Organization. Cli n. Pathol.1978;31:395414. 81. Friedman SL, Roll FJ, Boyles J, Bissell DM. Hepatic lipocytes: the principal collagenproducing cells of normal rat liver. Proc Natl Acad Sci USA. 1985;82(24):8681-5. 82. Verkhratrsky A. Calcium and cell deat h. Carafoli E and Brini M (eds.), pp. 465, Springer, Netherlands. Calcium Signalling and Disease; (2007). 83. Ferri K, Kroemer G. Organelle-specific initi ation of cell death pa thways. Nature Cell Biology 2001;3. 84. Nakagawa. Caspase-12 mediates endopl asmicreticulum-specific apoptosis andcytotoxicity by amyloid-beta. Let to Nature. 2000;403. 85. Friedman S, Arthur M. Ac tivation of cultured rat hepatic lipocytes by Kupffer cell conditioned medium: direct enhancement of matrix synthesis and stimulation of cell proliferation via induction of platelet-derived growth factor receptors. J. Clin. Invest. 1989;84:1780-1785. 86. Li D, Friedman S. Liver Fibrogenesis and the ro le of hepatic stellate cells: new insights and prospects for therapy. Gastroen terology and Hepatology 1999;14(7):618-33.

PAGE 100

100 87. Tsukada S, Westwick JK, Ikejima K, Sato N, Rippe RA. SMAD and p38 MAPK signaling pathways independently regulate alpha1 (I) collagen gene expression in unstimulated and transforming growth factor-betastimulated hepatic ste llate cells. J Biol Chem. 2005;280(11):10055-64. 88. Gressner OA, Lahme B, Demirci I, Gressner AM, Weiskirchen RJ. Diff erential effects of TGF-beta on connective tissue growth factor (CTGF/CCN2) expression in hepatic stellate cells and hepatocytes. Hepatology 2007;47(5):699-710. 89. Bissell DM. Chronic liver injury, TGF-b, and cancer. Experimental and Molecular Medicine 2001;33:4:179-190. 90. Blom IE, Goldschmeding R, Leask A.Bork P. Matrix Gene regulation of connective tissue growth factor: new targets for an tifibrotic therapy? Biol. 2002;21(6):473-82. 91. Badr MZ, Belinsky SA, Kauffman FC, and T hurman RG. Mechanism of hepatotoxicity to periportal regions of the liver lobule due to allyl alcohol: ro le of oxygen and lipid peroxidation. J Pharmacol Exp Ther 1986;238:1138-1142. 92. Boyland E and Chasseaud LF. Enzyme-catal ysed conjugations of glutathione with unsaturated compounds. Biochem. J. 1967;104:95. 93. Serafini-Cessi F. Conversion of allyl alcohol into acrolei n by rat liver. Biochem. J. 1972;128:1103-1107. 94. Jung SA, Chung YH, Park NH, Lee SS, Kim JA, Yang SH, Song IH, Lee YS, Suh DJ, Moon IH. Experimental model of hepatic fibr osis following repeated periportal necrosis induced by allylalcohol. Scand J Gastroenterol. 2000;35(9):969-75. 95. Ohno Y, Ormstad K, Ross D, Orrenius S. Mechanism of allyl alcohol toxicity and protective effects of low-molecular-weight thio ls studied with isolated rat hepatocytes. Toxicol Appl Pharmacol. 1985;78(2):169-79. 96. Hales CA, Barkin PW, Jung J, Trautman E, Lamborghini D, Herrig N and Burke J. Synthetic smoke with acrolein but not HCl produces pulmonary edema. J Appl Physiol. 1988;64:1121. 97. Ayer HE and Yeager DW. Irritants in cigarette smoke plumes. Am J Public Health 1982;72:1283-1299. 98. Kehrer JP, Biswal SS. The molecular effect s of acrolein. Toxicol. Sci. 2000;57: 6. 99. Vasiliou V, Pappa A, Estey T. Role of hu man aldehyde dehydrogenases in endobiotic and xenobiotic metabolism. Drug Metab Rev. 2004;36(2):279-99.

PAGE 101

101 100. Ellis EM. Reactive carbonyls and oxidative stre ss: potential for therapeutic intervention. Pharmacol Ther. 2007;115(1):13-24. 101. Morimoto K, and Takeshita T. Low Km aldehyde dehydrogenase (ALDH2) polymorphism, alcohol drinking behavior, and chromosome alterations in peripheral lymphocytes. Environmental Hea lth Perspectives 1996:104;3. 102. U.S. Public Health Service, Agency for Toxic Substances: Toxi cological profiles for carbon tetrachloride. 1989. 103. Raucy JL, Kraner JC, Lasker JM. Bio activation of halogenated hydrocarbons by cytochrome P4502E1. Crit Rev Toxicol. 1993;23(1):1-20. 104. Gruebele A, Zawaski K, Kaplan D, Novak RF. Cytochrome P4502E1and cytochrome P4502B1/2B2-catalyzed carbon tetrachloride me tabolism: effects on signal transduction as demonstrated by altered immediate-earl y (c-Fos and c-Jun) gene expression and nuclear AP-1 and NF-kappa B transcript ion factor levels. Drug Metab Dispos. 1996;24(1):15-22. 105. Gomez MI, Castro JA. Reacti on of trichloromethyl free ra dicals with deoxyribonucleic acid bases. Res Commun Chem Pa thol Pharmacol. 1981;32(1):147-53. 106. Connor HD, Lacagnin LB, Knecht KT, Thurman RG, Mason RP. Reaction of glutathione with a free radical metabolite of carbon te trachloride. Mol Pharmacol. 1990;37(3):44351. 107. Rechnagel RO and Glende EA. Carbon tetrachlori de hepatotoxicity: an example of lethal cleavage. CRC Crit Rev Tox.1973;2:263-297. 108. Farber E and El-Mofty SK. The biochemical pathology of liver cell necrosis. Am J Pathol. 1975;81:237-250. 109. Basu S. Carbon tetrachloride-induced lipid pe roxidation: eicosanoid formation and their regulation by antioxidant nutrien ts. Toxicology 2003;189(1-2):113-27. 110. Beddowes EJ, Faux S and Chipman K. Chloro form, carbon tetrachlor ide and glutathione depletion induce secondary genot oxicity in liver cells via ox idative stress. Toxicology 2003;187:101-115. 111. Smuckler E. Structural and functional cha nges in acute liver injury. Environmental Health Perspectives 1976;15:13-25. 112. Weber L, Boll M, Stampfl A. Hepatotoxicity and Mechanism of Ac tion of Haloalkanes: Carbon Tetrachloride as a Toxicological Model. Critical Reviews in Toxicology 2003;33(2):105-136.

PAGE 102

102 113. Ruchirawat M, Mostafa MH, Shank RC and We isburger EK. Inhibitory effects of carbon tetrachloride on diethylnitrosamine me tabolism and DNA alkylation. Carcinogenesis 1983;4:537-539. 114. Varela-Moreiras G, Alonso-Aperte E, Rubi o M, Gasso M, Deulofem R, Alvarez L, Caballeria J, Rodes J and Mato JM. Carbon tetrachloride-induced hepatic injury is associated with global DNA hypome thylation. Hepatology 1995;22:1310-3115. 115. Iredale JR Matrix turnover in fibrog enesis. Hepato-Gastroenterol.1996; 43:56-71. 116. Arthur MJ. Role of Ito cells in the degradati on of matrix in liver. J Gastroenterol Hepatol 1995;10:57-62. 117. Arii S, Mise M, Harada T, Furutani M, Ishi gami S, Niwano M, et al. Overexpression of matrix metalloproteinase 9 gene in hepato cellular carcinoma with invasive potential. Hepatology 1996;24:316-22. 118. Ashida K, Nakatsukasa H, Higashi T, Ohguc hi S, Hino N, Nouso K, et al. Cellular distribution of 92-kd type IV collagenas e/gelatinase B in human hepatocellular carcinoma. Am J Pathol. 1996;149:1803-11. 119. Hess J, Angel P and Schorpp-Kistner M. AP-1 subunits: quarrel and harmony among siblings. Journal of Cell Science 2004;117:5965-5973. 120. Gauldie J, Bonniaud P, Sime P, Ask K, Ko lb M. TGF-beta, Smad3 and the process of progressive fibrosis. Biochem Soc Trans. Aug 2007;35(4):661-4. 121. Burt DW, Law AS. Evolution of the transf orming growth factor-beta superfamily. Prog Growth Factor Res. 1992;5:99-118. 122. Roberts AB, Frolik CA, Anzano MA, Sporn MB. Transforming growth factors from neoplastic and nonneoplastic tissu es. Fed Proc. 1983;42(9):2621-6. 123. Brenner DA, Chojkier M. Acetaldehyde increases collagen gene transcription in cultured human fibroblasts. J Biol Chem. 1987;262(36):17690-5. 124. Daley SR, Ma J, Adams E, Cobbold SP, Wald mann H. A key role for TGF-beta signaling to T cells in the long-term acceptance of allografts. J Immunol. 2007;179(6):3648-54. 125. Ignotz RA, Endo T, Massagu J. Regulation of fibronectin and type I collagen mRNA levels by transforming growth factor-b eta. J Biol Chem. 1987;262(14):6443-6. 126. Carr BI, Hayashi I, Branum EL, Moses HL. Inhibition of DNA synthesis in rat hepatocytes by platelet-derived type beta transforming growth factor. Cancer Res. 1986;46(5):2330-4.

PAGE 103

103 127. Serra R. Transforming growth factor be ta. Encyclopedia of Life Sciences 2002. 128. Derynck, Jerett, Chen, Eaton, Bell, Assoia n, Roberts, Sporn and Goeddel. Human transforming growth factor complementary DNA sequence and expression in normal and transformed cells. Nature 1985;316:701-705. 129. Taipale J, Saharinen J, Keski-Oja J. Extrace llular matrix-associated transforming growth factor-beta: role in cance r cell growth and invasion. Adv Cancer Res. 1998;75:87-134. 130. O'Connor-McCourt MD, Wakefield LM. Latent transforming growth factor-beta in serum. A specific complex with alpha 2-macroglobulin. J Biol Chem. 1987;262(29):14090-9. 131. Wakefield LM, Winokur TS, Hollands RS, Ch ristopherson K, Levinson AD, Sporn MB. Recombinant latent transforming growth factor beta 1 has a l onger plasma half-life in rats than active transforming growth factor beta 1, and a differe nt tissue distribution. J Clin Invest. 1990;86(6):1976-84. 132. Massague J. TGF-beta signal transd uction. Annu Rev Biochem. 1998;67:753-91. 133. Assoian RK, Grotendorst GR, Miller DM, and Sporn MB. Cellular transformation by coordinated action of three peptide growth factors from human platelets. Nature 1984;309:804-806. 134. Eickelberg O, Khler E, Reichenberger F, Bertschin S, Woodtli T, Erne P, Perruchoud AP, Roth M. Extracellular matrix depositi on by primary human lung fibroblasts in response to TGF-beta1 and TGF-be ta3. Am J Pathol. 1999;276:L814-24. 135. Chen H, Li D, Roberts GJ, Saldeen T, Meht a JL. Eicosapentanoic ac id inhibits hypoxiareoxygenation-induced injury by attenuating upregulation of MMP -1 in adult rat myocytes. Cardiovasc Res. 2003;1:59(1):7-13. 136. Strange J, Roberts RA. Reduced expression of mature TGF beta 1 correlates with the suppression of rat hepatocyte apoptosis by th e peroxisome proliferator, nafenopin. Mutat Res. 1996;372(1):107-13. 137. Russell WE, Coffey RJ, Ouellette AJ, Moses HL Type beta transforming growth factor reversibly inhibits the early proliferative response to partial hepatectomy in the rat. Proc Natl Acad Sci USA. 1988;85(14):5126-30. 138. Wickert L, Chatain N, Kruschinsky K, Gre ssner AM. Glucocorticoids activate TGF-beta induced PAI-1 and CTGF expression in rat hepatocytes. Comp Hepatol. 2007;2(6):5. 139. Yata Y, Gotwals P, Koteliansky V, Rockey DC. Dose-dependent inhibition of hepatic fibrosis in mice by a TGF-beta soluble recep tor: implications for antifibrotic therapy. Hepatology. 2002;35(5):1022-30.

PAGE 104

104 140. Daniels JT, Schultz GS, Blalock TD, Garrett Q, Grotendorst GR, Dean NM, and Khaw PT. Mediation of transforming growth facto r-beta(1)-stimulated matrix contraction by fibroblasts: a role for connectiv e tissue growth factor in c ontractile scarring. Am J Pathol 2003;163:2043-2052. 141. Grotendorst GR, Okochi H, Hayashi N. A novel transforming growth factor beta response element controls the expression of the connective tissue growth factor gene. Cell Growth Differ. 1996;7(4):469-80. 142. Chen Y, Blom IE, Sa S, Goldschmeding R, Abraham DJ, Leask A. CTGF expression in mesangial cells: involvement of SMADs MAP kinase, and PKC. Kidney Int. 2002;62(4):1149-59. 143. Bork P. The modular architecture of a new family of growth regulators related to connective tissue growth factor. FEBS Lett. 1993;26:327(2):125-30. 144. Kim HS, Nagalla SR, Oh Y, Wilson E, Robert s CT Jr, Rosenfeld RG Identification of a family of low-affinity insulin-like gr owth factor binding proteins (IGFBPs): characterization of connective tissue grow th factor as a member of the IGFBP superfamily. Proc Natl Acad Sci USA. 1997;94(24):12981-6. 145. Voorberg J, Fontijn R, Calafat J, Janssen H, van Mourik JA, Pannekoek H. Assembly and routing of von Willebrand factor variants: the requirements for disulfide-linked dimerization reside within the carboxyterminal 151 amino acids. J Cell Biol. 1991;113(1):195-205. 146. Guo NH, Krutzsch HC, Ngre E, Vogel T, Blake DA, Roberts DD. Heparinand sulfatide-binding peptides from the type I repeats of human thrombospondin promote melanoma cell adhesion. Proc Natl Acad Sci USA. 1992;89(7):3040-4. 147. Brigstock DR. The connective tissue growth factor/cysteine-rich 61/nephroblastoma overexpressed (CCN) family. Endocr Rev. 1999;20:189-206. 148. Pi L, Ding X, Jorgensen M, Pan JJ, Oh SH, Pintilie DG, Brown AR, Song WY, and Petersen BE. Connective Tissue Growth Fact or with a Novel Fibronectin Binding Site Promotes Cell Adhesion and Migration Du ring Rat Oval Cell Activation. Hepatology epub. 149. Frazier K, Williams S, Kothapalli D, Kl apper H, Grotendorst GR. Stimulation of fibroblast cell growth, matrix production, and granulation tissue fo rmation by connective tissue growth factor. J Inve st Dermatol. 1996;107(3):404-11. 150. Kireeva ML, Latinki BV, Kolesnikova TV, Chen CC, Yang GP, Abler AS, Lau LF. Cyr61 and Fisp12 are both ECM-associated si gnaling molecules: activities, metabolism, and localization during development. Exp Cell Res. 1997;233(1):63-77.

PAGE 105

105 151. Wunderlich K, Senn BC, Todesco L, Flamme r J, Meyer P. Regul ation of connective tissue growth factor gene expression in re tinal vascular endothelia l cells by angiogenic growth factors. Graefes Arch C lin Exp Ophthalmol. 2000;238(11):910-5. 152. Pendurthi UR, Allen KE, Ezban M, Rao LV. Factor VIIa and thrombin induce the expression of Cyr61 and connective tissue grow th factor, extracellular matrix signaling proteins that could act as possible downstream mediators in factor VIIa x tissue factorinduced signal transduction. J Biol Chem. 2000 12;275(19):14632-41. 153. Chambers RC, Laurent GJ. Coagulation cas cade proteases and tissue fibrosis. Biochem Soc Trans. 2002;30(2):194-200. 154. Hahn A, Heusinger-Ribeiro J, Lanz T, Zenkel S, Goppelt-Struebe M. Induction of connective tissue growth factor by activation of heptahelical receptors. Modulation by Rho proteins and the actin cytoskel eton. J Biol Chem. 2000;275(48):37429-35. 155. Abraham DJ, Shiwen X, Black CM, Sa S, Xu Y, Leask A. Tumor necrosis factor alpha suppresses the induction of connective tissu e growth factor by transforming growth factor-beta in normal and scleroderma fibr oblasts. J Biol Chem. 2000;275(20):15220-5. 156. Bradham DM, Igarashi A, Potter RL, and Grotendorst GR. Conn ective tissue growth factor: a cysteine-rich mitogen secreted by human vascular endot helial cells is related to the SRC-induced immediate early gene product CEF-10. J Cell Biol. 1991;114:12851294. 157. Nishida T, Nakanishi T, Asano M, Shimo T, Takigawa M. Effects of CTGF/Hcs24, a hypertrophic chondrocyte-specif ic gene product, on the pro liferation and differentiation of osteoblastic cells in vitro. J Cell Physiol. 2000;184(2):197-206. 158. Wandji SA, Gadsby JE, Barber JA, Hamm ond JM. Messenger ribonucleic acids for MAC25 and connective tissue growth factor (CTGF) are inversely regulated during folliculogenesis and early luteogenesis. Endocrinology 2000;141(7):2648-57. 159. Igarashi A, Nashiro K, Kikuchi K, Sato S, Ihn H, Fujimoto M, Grotendorst GR, Takehara K Connective tissue growth factor gene expr ession in tissue secti ons from localized scleroderma, keloid, and other fibrotic skin disorders. J Invest Dermatol. 1996;106(4):729-33. 160. Frazier KS, Grotendorst GR. Expression of connective tis sue growth factor mRNA in the fibrous stroma of mammary tumors. In t J Biochem Cell Biol. 1997;29(1):153-61. 161. Graness A, Poli V, Goppelt-Struebe M. STAT3-independent inhibition of lysophosphatidic acid-mediated upregulation of connective tissue grow th factor (CTGF) by cucurbitacin I. Biochem Phar macol. Jun 2006;72(1):32-41.

PAGE 106

106 162. Ujike K, Shinji T, Hirasaki S, Shiraha H, Nakamura M, Tsuji T, Koide N. Kinetics of expression of connective tissue growth factor gene during liver rege neration after partial hepatectomy and D-galactosamine-induced liv er injury in rats. Biochem Biophys Res Commun. 2000;277(2):448-54. 163. Paradis V, Dargere D, Bonvoust F, et al. Effects and regulation of connective tissue growth factor on hepatic stella te cells. Lab Invest 2002;82:767. 164. Hayashi N, Kakimuma T, Soma Y, Grotendorst GR, Tamaki K, Harada M, and Igarashi A. Connective tissue growth factor is directly related to liver fibrosis. Hepatogastroenterology 2002;49:133-135. 165. Crean JK, Finlay D, Murphy M, Moss C, G odson C, Martin F, Brady HR. The role of p42/44 MAPK and protein kinase B in c onnective tissue growth factor induced extracellular matrix prot ein production, cell migration, and actin cytoskeletal rearrangement in human mesangial ce lls. J Biol Chem. 2002;277(46):44187-94. 166. Segarini PR, Nesbitt JE, Li D, Hays LG, Ya tes JR 3rd, Carmichael DF. The low density lipoprotein receptor-related protein/alpha2-m acroglobulin receptor is a receptor for connective tissue growth factor J Biol Chem. 2001;276(44):40659-67. 167. Shupe T, Sell S. Low hepatic glutathione S-transferase and increased hepatic DNA adduction contribute to increase d tumorigenicity of aflatoxin B1 in newborn and partially hepatectomized mice. Toxicol Lett. 2004 14;148(1-2):1-9. 168. Kinnman N, Housset C. Peribilia ry myofibroblasts in biliar y type liver fibrosis. Front Biosci. 2002;7:496-503.

PAGE 107

107 BIOSKETCH Alicia Renae Brown was born in 1983 to Gary Ray and Julie Ann (Darr) Brown of Alton, IL. She was preceded in birth by her sisters Christin a Rae, 4 years and less than a day older, and Andrea Ann by 5 minutes, her fraternal twin sister They grew up just outside of Brighton, IL, a small town about an hour north of St. Louis, MO, and attended Brighton Elementary then Southwestern High School. Alicia advanced to the University of Illinois at Urbana-Champaign in Animal Sciences with the intent of going into research. While there, she worked in numerous labs in areas including Re productive Biology and Endocri nology, Swine Immunology, Maize Genetics and Nueroimmunology. She has worked direc tly with respectable professors such as Drs. Robert Lambert, Humphrey Yao, Janice Bahr and Edward Roy. Her senior year at the University of Illinois she applied to several sc hools to obtain a Master's degree in Biology and she chose the University of Florida based on its top-ranking medical school. At the University of Florida, she decided to work in the laboratory of Bryon Petersen as the laboratory manager while she continued her school work.