<%BANNER%>

Skeletal Effects of Teriparatide in Glucocorticoid-Treated Mice

Permanent Link: http://ufdc.ufl.edu/UFE0020972/00001

Material Information

Title: Skeletal Effects of Teriparatide in Glucocorticoid-Treated Mice
Physical Description: 1 online resource (121 p.)
Language: english
Creator: Howe, Kathleen S
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: animal, glucocorticoid, osteoporosis, parathyroid, prednisolone, pth, rodent, teriparatide
Applied Physiology and Kinesiology -- Dissertations, Academic -- UF
Genre: Health and Human Performance thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Synthetic analogs of glucocorticoid (GC) drugs are widely used in treating many inflammatory diseases and conditions and are also used to suppress the immune system in solid organ transplant recipients. However GCs have serious side effects including osteoporosis and bone fractures. We conducted a randomized, prospective investigation of the effects of teriparatide in treating GC-induced osteopenia in mice and examined the extent and character of bone recovery in the distal femur and lumbar spine after exposure to GC when there was No subsequent treatment with teriparatide Simultaneous GC and teriparatide administration over the entire course of treatment Delayed administration of teriparatide Seven month old male Swiss Webster mice received prednisolone (2.1 mg/kg/d), teriparatide (40ug/kg/d), or vehicle to determine changes in bone structure and turnover after a 4- or 8-week (6 d/wk) treatment regimen. We injected flurochrome markers (declomycin and calcein) before sacrifice and harvested femurs and lumbar vertebrae to assess bone response. Bone samples were analyzed using histomorphometry and microCT and both techniques showed the same trends. We found GCs suppressed bone turnover but not necessarily bone volume and that teriparatide, a bone anabolic agent, effectively increased bone turnover and inhibited bone changes resulting from GC exposure. The effects of teriparatide were rapid and relative changes were greater in the distal femur than in the lumbar spine. Four and eight weeks of teriparatide significantly increased both the osteoclast surface (Oc.S) and the osteoblast surface (Ob.S), resulting in significant increases in mineralizing surface and mineral apposition rate. Increased Ob.S and Oc.S indicated the increased turnover seen with teriparatide favored bone formation. We also detected a residual effect of GC on bone evidenced by lack of increased bone formation despite increased osteoblastic activity after GC treatment was discontinued. The underlying goal of our study was to demonstrate the efficacy of using PTH to prevent the adverse effects of GCs on bone in mice, as a prelude to studies in humans.
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.
Statement of Responsibility: by Kathleen S Howe.
Thesis: Thesis (Ph.D.)--University of Florida, 2007.
Local: Adviser: Braith, Randy W.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0020972:00001

Permanent Link: http://ufdc.ufl.edu/UFE0020972/00001

Material Information

Title: Skeletal Effects of Teriparatide in Glucocorticoid-Treated Mice
Physical Description: 1 online resource (121 p.)
Language: english
Creator: Howe, Kathleen S
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: animal, glucocorticoid, osteoporosis, parathyroid, prednisolone, pth, rodent, teriparatide
Applied Physiology and Kinesiology -- Dissertations, Academic -- UF
Genre: Health and Human Performance thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Synthetic analogs of glucocorticoid (GC) drugs are widely used in treating many inflammatory diseases and conditions and are also used to suppress the immune system in solid organ transplant recipients. However GCs have serious side effects including osteoporosis and bone fractures. We conducted a randomized, prospective investigation of the effects of teriparatide in treating GC-induced osteopenia in mice and examined the extent and character of bone recovery in the distal femur and lumbar spine after exposure to GC when there was No subsequent treatment with teriparatide Simultaneous GC and teriparatide administration over the entire course of treatment Delayed administration of teriparatide Seven month old male Swiss Webster mice received prednisolone (2.1 mg/kg/d), teriparatide (40ug/kg/d), or vehicle to determine changes in bone structure and turnover after a 4- or 8-week (6 d/wk) treatment regimen. We injected flurochrome markers (declomycin and calcein) before sacrifice and harvested femurs and lumbar vertebrae to assess bone response. Bone samples were analyzed using histomorphometry and microCT and both techniques showed the same trends. We found GCs suppressed bone turnover but not necessarily bone volume and that teriparatide, a bone anabolic agent, effectively increased bone turnover and inhibited bone changes resulting from GC exposure. The effects of teriparatide were rapid and relative changes were greater in the distal femur than in the lumbar spine. Four and eight weeks of teriparatide significantly increased both the osteoclast surface (Oc.S) and the osteoblast surface (Ob.S), resulting in significant increases in mineralizing surface and mineral apposition rate. Increased Ob.S and Oc.S indicated the increased turnover seen with teriparatide favored bone formation. We also detected a residual effect of GC on bone evidenced by lack of increased bone formation despite increased osteoblastic activity after GC treatment was discontinued. The underlying goal of our study was to demonstrate the efficacy of using PTH to prevent the adverse effects of GCs on bone in mice, as a prelude to studies in humans.
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.
Statement of Responsibility: by Kathleen S Howe.
Thesis: Thesis (Ph.D.)--University of Florida, 2007.
Local: Adviser: Braith, Randy W.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0020972:00001


This item has the following downloads:


Full Text





SKELETAL EFFECTS OF TERIPARATIDE INT GLUCOCORTICOID-TREATED MICE


By

KATHLEEN S. HOWE
















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

UNIVERSITY OF FLORIDA

2007







































O 2007 Kathleen S. Howe


































To my mother and father. My father' s recent passing brought great sadness to our family, but I
know he would be proud of this accomplishment. My mother has been extremely supportive and
her encouragement was instrumental in making this dissertation a reality.









ACKNOWLEDGMENTS

No proj ect of this size occurs without a great deal of support and encouragement. First,

and most importantly, I wish to thank the many individuals who directly contributed to this

research study. Jodi Long was involved in every aspect of this study. Her enthusiasm,

knowledge, and energy contributed immensely to this proj ect. I also owe a debt of gratitude to

Dr. Randy Braith for his guidance and support throughout this proj ect and my tenure at this

university. His extensive knowledge and support were invaluable to me as I designed and

conducted this study. His comments, suggestions, and editorial talents helped bring the study

results into sharper focus.

I owe a particular debt of gratitude to Dr. Tom Wronski who helped design this study,

allowed me to use equipment in his lab, and helped me to interpret the results. He is the

foremost expert on bone biology at the University of Florida and I profited greatly from his

knowledge. I would also like to thank Molly Altman and Sally Vanegas who prepared and

analyzed bone samples for histomorphometric analysis and Ignacio Aguirre who was always

willing to answer my questions. Their knowledge and professionalism helped make this project

a success.

I would also like to thank Dr. Russell Turner and Dr. Urszula Iwaniec. They graciously

permitted me access to their equipment and expertise. Their enthusiasm for this proj ect and

willingness to teach me about bone research has helped my professional development

immensely. I have also benefited greatly from my committee's guidance and thank both Dr.

Stephan Dodd and Dr. Scott Powers. Their insightful comments and probing questions helped

make this a better study. More than that, though, over the past seven years they have provided

encouragement and helped shape my professional development. I also owe a debt of gratitude to

the wonderful staff at Animal Care Services. They provided outstanding support, training, and









services to this project. Working with them was truly a pleasure. I would also like to thank

some of my friends who supported me: Susan Smith who always had a kind word, Vij a Purs

who grudgingly accepted my research with animals, Ben Webster, who kept me smiling, and,

most of all Janet Degner who was always there with a word of encouragement when the going

got rough. I also want to thank members of the Applied Physiology and Kinesiology staff. Kim

Hatch and Candyce Hudson have provided expert advice and support over the years. Their

willingness to help students and their expertise in managing grants greatly eased the way for this

project. I would also like to thank James Milford and Susie Weldon. They are the collective

memory for the department and seem to always know how to make things happen. Their "can-

do" attitude and enthusiastic support was very much appreciated. Finally, I would like to thank

my mother who stood behind me unfailing through this arduous journey.












TABLE OF CONTENTS


page


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


LIST OF TABLES ............. ...... ___ ...............9....


LIST OF FIGURES ............. ...... ___ ...............11...


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


CHAPTER


1 INTRODUCTION ................. ...............14.......... ......


Study Purpose ................. ...............14.......... .....
Rationale for Study ................. ...............15................
Study Aims ................... ...............18..
Significance of the Study ................. ...............21................

2 MATERIALS AND METHODS .............. ...............23....


Back ground ................. ...............23.......... ......
A nim als................ .... .............2

Animal Housing Conditions .............. ...............24....
Study Group Assignment............... ...............2
Pharmacological Agents .............. ...............25....
Study Drugs. .............. ... ...............25.
Prednisolone Succinate. ............. ...............26.....

Teriparatide. ............. ...............26.....
Flurochrome markers............... ...............27
Demeclocycline. .............. ...............27....
Calcein............... ...............27
Anesthesia and Euthanasia .............. ...............28....

Bone Harvesting .............. ...............28....
Study Measures............... ...............29
Anthropomorphic Measures .............. ...............29....
Hi stomorphometry ................. .....___ ...............29.......
Static bone measurements .............. ...............30....

Dynamic bone measurements............... .............3
MicroCT .............. ...............3 2....
F em ur .............. ...............3 2....
Vertebrae .............. ...............3 3....
Statistical Analysis............... ...............34















3 LITERATURE REVIEW ................. ...............36...... .....


Bone Biology ....._._. ................ ..........._..........3
Structure of Bone............... ...............36.
Bone Cells .............. ...............38....
Osteoclasts ............._. ...._... ...............38....
O steoblasts .............. ...............39....

O steocytes .............. ...............39....
Bone Lining Cells............... ...............40.
Bone Rem odeling ........._..._.._ ................... .............. .. .... ..........4
Remodeling Balance The RANKL/OPG/RANK Axis............... ...............43..
Effects of Glucocorticoid Drugs and Teriparatide on Bone .................. ................4

Systemic Effects of Glucocorticoid Drugs ....._._._ .......__. ...._._ ...........4
Direct Effects of Glucocorticoids ................. ...............45................

Indirect Effects of Glucocorticoid Drugs .............. ...............46....
Decreased intestinal absorption of calcium ....._____ ...... ..___ .. ........_......46
Increased renal elimination of calcium .............. ...............47....

Antagonistic action on gonadal functions .............. ...............47....
Increased sensitivity to PTH .............. ... ... ...............48
Bone loss in response to glucocorticoid treatment ................. ................ ......... .48
Parathyroid Horm one................. ... .. ...... ..... ..........4
Dual Nature of PTH: Continuous versus Intermittent Administration...........................4

Parathyroid Hormone (PTH 1-84) ........._....._ ...._.._....._._ ....... ....4

Teriparatide (PTH 1-34) ........._... ......._ ...............51..
Mechanisms of Action .............. ......__ .... ...............52.
Studies of Glucocorticoid-Induced Bone Loss in Mice............... .. .. ..... ...........5

Validity of the Mouse as a Model of Glucocorticoid-Induced Bone Loss ........._..........53
Glucocorticoid-Induced Bone Loss in Mice............... ...............53..

Teriparatide Treatment in Mice............... ...............54..
Further Considerations............... .............5


4 RE SULT S .............. ...............56....


M easurement Design .............. ...............56....
Anthropomorphic Measures .............. ...............57....
Bone M easures................... ....................5
Measurements of the Lumbar Vertebrae ....__ ................ ...............58.....
Bone Volume/Total Volume .............. ...............59....
Trab ecular Numb er ................. ............... ...............61.....
Trabecular Width/Trabecular Thickness .............. ...............63....

Trabecular Separation............... ...............6
Measurements of the Distal Femur ........._..._.._ ...._._. ...............67...
Bone Volume ........._..._.._ ...._._. ...............68.....
Trab ecular Numb er ................... ........_..._. ...............71....
Trabecular Thickness/Trabecular Width .........._..... ....._.. ........_ ............7











Trabecular Separation................... ....... .........7
Osteoblast Surface and Osteoclast Surface .............. ...............76....
Dynamic Measures of Bone Formation in the Distal Femur ................. .......................79
M ineralizing Surface .............. ...............79....
M ineral Apposition Rate .............. .................. ..........8
Bone Formation Rate/Bone Surface (BFR/B S) ................. ...............82...............
Mid-Shaft Cortical Bone Data and Significant Changes ................. ................. ........ 84

5 DI SCUS SSION ................. ...............86................


Glucocorticoid Drugs Suppressed Bone Formation But Did Not Affect Bone Volume........86
Anabolic Effects of PTH Prevented the Inhibitory Changes Associated with
Glucocorticoid Drugs ............ ..... .._ ...............89...
PTH Increases Bone Mass Quickly .............. .... .......... ....... .......9
Residual Effects of Glucocorticoid Drugs are Apparent During Natural Recovery .............92
PTH Treatment After Glucocorticoid Use Was More Effective than Natural Recovery......93
Age-Related Effects on Bone Mass ................. ... .......... .... ...... .... ................9
Prophylactic Value of Concurrent Treatment with Glucocorticoid Drugs and PTH..............96
Site Specificity of GC and PTH Treatment ................ ...............97........... ..
Conclusions............... ..............9
Clinical Applications .............. ...............98....
Study Limitations............... ...............9
Future Directions .............. ...............100....


APPENDIX

A SUMMARY OF BONE MEASUREMENTS ................ ...............101........... ...

B SUMMARY OF SELECTED STUDIES IN MICE ................. ...._.. ................ ...10

LIST OF REFERENCES ...._.. ................. ......._.. .........11

BIOGRAPHICAL SKETCH ....._.. ................ .........__..........12










LIST OF TABLES


Table page

3-1 Common Effects of Glucocorticoid Therapy ................. ........._.. ...... 45_._. ...

4-1 Experimental Groups and Description of Treatments. ............. ...............56.....

4-2 Mean Animal Weights by Group. .............. ...............57....

4-3 Mean Femur Lengths by Group. .............. ...............57....

4-4 Summary of Histomorphometric Analysis of LV3 by Group. ................... ...............5

4-5 Summary of MicroCT Analysis of LV2 by Group. ........._.._.. ...._... ......_.._.......59

4-6 Significant Changes in Lumbar Vertebra L3 Bone Volume by Group using
Histomorphometry. ............. ...............60.....

4-7 Significant Changes in Lumbar Vertebra L2 Bone Volume by Group using MicroCT....61

4-8 Significant Changes in Lumbar Vertebra L3 Trabecular Width by Group using
Histomorphometry. ............. ...............64.....

4-9 Significant Changes in Lumbar Vertebra L2 Trabecular Thickness by Group using
M icroCT ................. ...............65.................

4-11 Summary of MicroCT Analysis of the Distal Femur by Group. ................ .................. 68

4-12 Significant Changes in Distal Femur Bone Volume by Group using
Histomorphometry. ............. ...............69.....

4-13 Significant Changes in Distal Femur Bone Volume by Group using MicroCT. ...............70

4-15 Significant Changes in Distal Femur Trabecular Number by Group using MicroCT.......72

4-16 Significant Changes in Distal Femur Trabecular Width by Group using
Histomorphometry. ............. ...............73.....

4-17 Significant Changes in Distal Femur Trabecular Thickness by Group using MicroCT....74

4-18 Significant Changes in Distal Femur Trabecular Separation by Group using
Histomorphometry. ............. ...............75.....

4-19 Significant Changes in Distal Femur Trabecular Separation by Group Using
M icroCT ................. ...............76.................

4-20 Significant Changes in Distal Femur Osteoblast Surface and Osteoclast Surface by
G roup. ............. ...............77.....










4-21 Significant Changes in Distal Femur Mineralizing Surface by Group .............. ...............80

4-22 Significant Changes in Distal Femur Mineral Apposition Rate by Group. .......................81

4-23 Significant Changes in Distal Femur Bone Formation Rate by Group using
Histomorphometry. ............. ...............83.....

4-24 Summary of Histomorphometric Analysis of Femur Mid-Shaft by Group. ................... ...84

4-25 Summary of Significant Changes in Femur Mid-Shaft Cortical Bone Thickness by
Group using MicroCT ................. ...............84........... ....

A-1 Summary of Significant Changes in Lumbar Vertebrae L3 based on
Histomorphometry. ............. ...............101....

A-2 Summary of Significant Changes in Lumbar Vertebrae L2 based on MicroCT. ............102

A-3 Summary of Significant Changes in the Distal Femur based on Histomorphometry ......103

A-4 Summary of Significant Changes in the Distal Femur based on MicroCT .....................104

A-5 Percent Changes in Osteoclast and Osteoblast Surfaces in the Distal Femur. .................1 05

A-6 Percent Changes in Dynamic Bone Formation Parameters in the Distal Femur. ............106

B-1 Studies Of Glucocorticoid-Induced Bone Loss In Mice. ....._____ ...... ..__ ............108

B-2 Studies using Teriparatide in Mice. ..........._._ ....._._ ...............110..










LIST OF FIGURES


Figure page

2-1 Study Groups, Treatments and Timelines. ....._.._._ ... .....___ .....__ ...........2

4- 1 Lumbar Vertebra L3 Bone Volume/Total Volume by Group using
Histomorphometry. ............. ...............60.....

4-2 Lumbar Vertebra L2 Bone Volume/Total Volume by Group using MicroCT. ...............61

4-3 Lumbar Vertebra L3 Trabecular Number by Group using Histomorphometry .................62

4-4 Lumbar Vertebra L2 Trabecular Number by Group using MicroCT ............... ...............62

4-5 Lumbar Vertebra L3 Trabecular Width by Group using Histomorphometry. .................64

4-6 Lumbar Vertebra L2 Trabecular Thickness by Group using MicroCT. ............................65

4-7 Lumbar Vertebra L3 Trabecular Separation by Group using Histomorphometry. ..........66

4-8 Lumbar Vertebra L2 Trabecular Separation by Group using MicroCT. ...........................67

4-9 Distal Femur Bone Volume/Total Volume by Group using Histomorphometry. ...........69

4-10 Distal Femur Bone Volume/Total Volume by Group using MicroCT. .............................70

4-11 Distal Femur Trabecular Number by Group using Histomorphometry ................... ..........71

4-12 Distal Femur Trabecular Number by Group using MicroCT. ............. .....................7

4-13 Distal Femur Trabecular Thickness by Group using Histomorphometry. .......................73

4-14 Distal Femur Trabecular Thickness by Group using MicroCT. ........... .....................74

4-15 Distal Femur Trabecular Separation by Group using Histomorphometry. ......................75

4-16 Distal Femur Trabecular Separation by Group using MicroCT. .......... ....................76

4-17 Distal Femur Osteoblast Surface by Group using Histomorphometry. ........... ................78

4-19 Distal Femur Mineralizing Surface by Group using Histomorphometry. .........................80

4-20 Distal Femur Mineral Apposition Rate by Group using Histomorphometry. ...................81

4-21 Distal Femur Bone Formation Rate/Bone Surface by Group using
Histomorphometry. ............. ...............83.....

4-22 Mid- Shaft F emur Corti cal Thi ckne ss by Group using Mi croCT ................. ................. .8 5









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

SKELETAL EFFECTS OF TERIPARATIDE INT GLUCOCORTICOID-TREATED MICE


By

Kathleen S. Howe

August 2007

Chair: Randy W. Braith
Major : Health and Human Performance

Synthetic analogs of glucocorticoid (GC) drugs are widely used in treating many

inflammatory diseases and conditions and are also used to suppress the immune system in solid

organ transplant recipients. However GCs have serious side effects including osteoporosis and

bone fractures. We conducted a randomized, prospective investigation of the effects of

teriparatide in treating GC-induced osteopenia in mice and examined the extent and character of

bone recovery in the distal femur and lumbar spine after exposure to GC when there was

* no subsequent treatment with teriparatide;
* simultaneous GC and teriparatide administration over the entire course of treatment;
* delayed administration of teriparatide


Seven month old male Swiss Webster mice received prednisolone (2.1 mg/kg/d), teriparatide

(40ug/kg/d), or vehicle to determine changes in bone structure and turnover after a 4- or 8-week

(6 d/wk) treatment regimen. We injected flurochrome markers (declomycin and calcein) before

sacrifice and harvested femurs and lumbar vertebrae to assess bone response. Bone samples

were analyzed using histomorphometry and microCT and both techniques showed the same

trends.









We found GCs suppressed bone turnover but not necessarily bone volume and that

teriparatide, a bone anabolic agent, effectively increased bone turnover and inhibited bone

changes resulting from GC exposure. The effects of teriparatide were rapid and relative changes

were greater in the distal femur than in the lumbar spine. Four and eight weeks of teriparatide

significantly increased both the osteoclast surface (Oc. S) and the osteoblast surface (Ob.S),

resulting in significant increases in mineralizing surface and mineral apposition rate. Increased

Ob.S and Oc.S indicated the increased turnover seen with teriparatide favored bone formation.

We also detected a residual effect of GC on bone evidenced by lack of increased bone formation

despite increased osteoblastic activity after GC treatment was discontinued. The underlying goal

of our study was to demonstrate the efficacy of using PTH to prevent the adverse effects of GCs

on bone in mice, as a prelude to studies in humans.









CHAPTER 1
INTTRODUCTION

Glucocorticoid-induced bone loss is the leading cause of secondary osteoporosis (1). The

negative effects of glucocorticoid (GC) drugs on the skeletal system are well established but

there is currently no consensus on the best way to prevent and/or treat the associated bone loss.

We hypothesized that teriparatide, currently the only anabolic agent approved for the treatment

of established osteoporosis, would be capable of reversing and/or preventing glucocorticoid-

induced bone loss. There are a number of patient populations that could benefit from such

treatment. Chronic lung disease, rheumatic diseases, and gastrointestinal diseases often involve

prophylactic GC use. Patients awaiting solid organ transplants would also benefit, since some

transplant centers consider antecedent osteoporosis a contraindication for transplant surgery

because immunosuppressant regimens including GC cause rapid bone loss after transplantation.

There have been studies examining the effects of GC use in mice (2-5) but no studies have

examined the combined use of GC and teriparatide in an animal model. Furthermore, there has

only been one study documenting the use of teriparatide in humans exposed to long-term GC

therapy (6).

The limited use of teriparatide in humans treated with GC prompted us to select a mouse

model for this study. We used the Swiss Webster strain of mice because they have significant

levels of cancellous bone in the femur and spine (7), have previously shown loss of bone in

response to GC treatment (4,5), and an anabolic response to teriparatide (8). A mouse model

allowed us to simulate chronic GC use.

Study Purpose

The purpose of this study was to determine the extent/character of bone recovery following

exposure to GC drugs when there was










* No subsequent treatment with teriparatide
* Simultaneous GC and teriparatide administration over the entire course of treatment
* Delayed administration of teriparatide

We used microCT and histomorphometric techniques to evaluate bone mass, bone

resorption, bone formation, and microarchitectural endpoints such as trabecular number,

thickness, spacing, and connectivity density to determine whether teriparatide improved bone

mass and architecture following exposure to GC. This preliminary study, designed to assess the

effects of teriparatide on glucocorticoid-induced osteopenia* in mice, was the first step in a

research sequence that will help define new treatment options to improve the quality of life for

clinical populations exposed to long-term GC therapy and those facing transplant surgeries.

Rationale for Study

Synthetic analogs of GC are a widely used class of drugs that have proven effective in

many inflammatory diseases and conditions, including asthma, Chronic Obstructive Pulmonary

Disease, rheumatoid arthritis, Crohn's disease and lung diseases such as cystic fibrosis. GC ana-

logs are also a key anti-rej section drug following solid organ transplantation (9,10). However,

GC drugs pose serious side effects to the patient. Osteoporosis, with resulting bone fractures, is

the most incapacitating sequelae of GC therapy. Bone is a dynamic, living tissue in which there

is a normal balance of bone formation and bone resorption. This balance of bone loss and gain

helps maintain a healthy skeletal structure capable of withstanding normal loads and stresses.

GC drugs disrupt the normal homeostasis of bone and rapidly lead to loss of bone mass and

increased fracture risk. GCs have a negative effect on both the hard outer layer of cortical bone

and the cancellous bone found next to the marrow. Although GC drugs affect both types of

bone, the most profound and rapid effects are seen in cancellous bone.

SOsteoporosis is a term based on T-scores established for human populations. As such, the term osteopenia rather
than osteoporosis is used to describe decreased bone mass in animals.









The relationship between long-term GC use and osteoporosis is well established and is

often referred to as Glucocorti coid-Induced Osteoporosi s (GIO) in humans (1 1 -14). Deleterious

effects on the bones are found in dosages often met or exceeded in the treatment of many

conditions. Dosages as small as 7.5 mg/day can result in a loss of spinal trabecular bone of 9.5%

in 5 months (15) indicating even low doses can cause significant loss of bone. Bone loss occurs

most rapidly in the first 6-12 months of treatment and appears to be dose and duration dependent

(15-20). Osteoporosis has been reported in 50% of patients exposed to long-term GC treatment

and spinal fractures occur at a rate 4-5 times that found in patients not treated with

glucocorticoids (12,15). Fracture rates among those taking the drugs for more than five years

approach 30% (21).

In solid organ transplant patients, significant loss of bone mass can be detected as early as

three months after transplantation (22). Bone mineral density losses average 5-15% during the

first year and 1-2% annually subsequently (16,23). The significant morbidity and mortality

associated with GIO makes its potential prevention or reversibility an important issue.

Currently there is no established method for preventing GIO. A variety of anti-resorptive

treatments have been tried, including calcium supplementation, bisphosphonate agents,

estrogenic and androgenic hormones, and calcitonin, but none of these has proven effective in

reversing the low bone formation that accompanies long-term GC use (6). We found that

calcium/vitamin D supplementation and nasal calcitonin can slow bone loss but is unable to

restore lost bone mass (24). Targeted resistance exercise and bisphosphonates have been shown

to prevent spinal bone loss in solid organ transplant patients but long-term compliance is

problematic (10,25,26). A recent study involving healthy postmenopausal women showed that a

combination of a bisphosphonate and a high impact exercise program increased bone mass more









effectively than bisphosphonate treatment alone (27). However, follow-up testing 15 months

after cessation of the intervention showed bone gains had not been maintained. If these gains

cannot be maintained in otherwise healthy populations, it is unlikely patients taking GC drugs

will be able to do so.

Teriparatide proved more efficacious in preventing and/or reversing GIO. Unlike the anti-

resorptive drugs such as bisphosphonates, which only slow bone loss, teriparatide has an

anabolic effect on bone, which may contribute to increased bone microarchitecture and strength.

Studies using teriparatide have shown increases in bone density at the femur neck and

particularly the lumbar spine (28-31) and indicate increases are greater with teriparatide than

with bisphosphonates (32,33). Fracture reduction has also been associated with the use of

teriparatide (29,3 1,34). A study of postmenopausal women with one to two preexisting non-

traumatic vertebral fractures showed that the vertebral fracture risk following teriparatide

treatment was reduced by two-thirds and the relative risk of non-vertebral fractures was reduced

by one-half (29). To date, there has only been one study examining the efficacy of teriparatide in

GIO in humans (6). However, the results of this study were confounded by a simultaneous use

of hormone replacement therapy, which acts as an anti-resorptive drug on bone. In that study,

the combination of estrogen and teriparatide resulted in significant bone density increases in the

axial skeleton but it is unclear whether teriparatide alone can overcome the deleterious effects of

GC treatment.

A number of studies have also shown that teriparatide stimulates bone formation, increases

cancellous bone volume, architecture, cortical width, and biomechanical properties of bone in

both mice (35) and humans (29). Teriparatide has been shown to increase cancellous bone and

bone mineral density particularly in the axial skeleton (8,35). Rodents are frequently used as a









model for osteoporosis research because they exhibit bone mass changes similar to humans when

exposed to many osteoporosis s-inducing stimuli (5). While the ovariectomized rat is the most

commonly used animal model for postmenopausal osteoporosis, the mouse may be a better

model for glucocorticoid-induced osteopenia (3,5) because researchers have found inconsistent

responses to GC exposure suggesting rats may be resistant to the effects of GC exposure (36-3 8).

Studies in mice which have achieved skeletal maturity (4,5) suggest the efficacy of the Swiss

Webster strain of mouse in a glucocorticoid-induced model of bone loss (2,5,7). Bone loss

patterns in mice exposed to GCs approximate human responses and the response to teriparatide

in studies suggests the process is similar in both humans and mice, leading us to choose this

animal for the study (3,5,39).

There is also a growing recognition that along with bone mineral density (BMD), bone

architecture should be examined to determine the true efficacy of a treatment (40,41). This study

was designed to use both histomorphometry and microCT techniques to determine bone

responses to GC treatment and whether there was any natural recovery following withdrawal of

that treatment. We also, for the first time, determined differing bone response to simultaneous

treatment with glucocorticoid and teriparatide (prevention) versus subsequent treatment with

teriparatide after bone loss has occurred (reversal).

Study Aims

Research Aim 1. To measure the changes in bone architecture and bone metabolism

resulting from teriparatide therapy following, or in conjunction with, GC administration in a

skeletally mature animal model.

Hypothesis 1. In a skeletally mature mouse, teriparatide will reverse bone loss caused by

4 or 8 weeks of prednisolone treatment.









Rationale: Mice experience dose-dependent loss of bone in response to GC treatment

over a threshold level of 1.4 mg/kg for as little as 27 days although evidence of increased

osteoclast numbers appears as early at as 10 days of treatment (4,5). This study used a dose of

2.1 mg/kg body weight, consistent with other studies (4,5). Previous studies using mice have

shown that chronic GC suppressed bone formation leading to bone loss in both axial and

appendicular skeletal sites (2-5). Researchers have noted increased bone resorption (4),

osteocyte apoptosis (5), and histomorphometric changes consistent with bone loss (2,3) in

response to GC treatment. Subsequent treatment with alendronate resulted in increased

osteoclast apoptosis and prevention of osteoblast apoptosis (4). However, although alendronate

slowed GC-induced bone loss, it could not prevent it (4).

Research Aim 2. To determine whether there are benefits to treating mice with

teriparatide as a prophylactic measure by comparing results when GCs and teriparatide are

administered together versus using teriparatide after glucocorticoid-induced bone loss has

already occurred.

Hypothesis 2. Starting teriparatide therapy at the same time as GC treatment will result in

less bone loss than starting teriparatide after glucocorticoid-induced osteopenia has developed.

Rationale: Numerous studies in humans and animals have shown that GC treatment has

negative effects on bones. GCs directly affect bone cells at least in part by increasing osteoclast

differentiation and activation levels (4,5), and increasing osteocyte and osteoblast apoptosis

(4,5,42,43). Teriparatide's anabolic effects on bone have the potential to slow or reverse these

effects. Patients using teriparatide after taking GCs for at least one year showed marked

increases in bone density at the spine although these results were less evident at the hip (6). We

believed a reversal of bone loss would also be seen in GC-treated mice also treated with










teriparatide. We believed treating animals with teriparatide after GC exposure would likely

attenuate bone loss and allow rebuilding of some microarchitectural features. Increased

resorption following GC use can cause loss of trabeculae and we did not expect teriparatide to

reverse this effect, since teriparatide can only build on existing bone. However, we believed the

use of teriparatide after GC exposure would reverse some of the damage and that we would see a

greater treatment effect when teriparatide treatment was started at the outset of GC exposure.

We believed beginning teriparatide treatment concurrently with GC would result in less overall

bone loss since teriparatide would offset the negative effects of the GC drugs.

Research Aim3. To determine the extent of unassisted recovery from GC therapy

compared with recovery using teriparatide.

Hypothesis 3. A therapy regimen consisting of 4 weeks of prednisolone use followed by 4

weeks of teriparatide treatment will result in increased bone mass and improved architecture

greater than will be seen from natural recovery after withdrawal of GC.

Rationale: We expected GC drugs to decrease bone mass and alter bone architecture. If

GC use was halted, we believed bone metabolic processes would return to baseline levels and

there would be a gradual restoration of at least some of the lost bone. We expected the natural

repair process to be less robust, however, than if teriparatide treatment had been initiated after

GC use. Teriparatide is highly anabolic and increases bone turnover and alters metabolism in

favor of bone formation. Although teriparatide would be unable to replace lost trabeculae we

expected to see increased bone formation on the surface of existing trabeculae.

Research Aim 4. Determine the degree to which skeletal responsiveness to teriparatide

would vary by site (femur versus lumbar vertebrae) based on differences in the prevalence of

trabecular bone at these sites.









Hypothesis 4. The deleterious effects of GCs and beneficial effects of teriparatide will be

seen first and most extensively in the vertebrae.

Rationale: Human studies indicate that the positive effects of teriparatide following GC

treatment appear first and to the greatest degree in the lumbar vertebrae (6). In the only study in

humans to date examining the effects of teriparatide in GIO, increases in bone density were first

detected in the spine (6). Changes in the hip were detected after 12 months of treatment,

although these increases in bone density were detected after teriparatide treatment had ended

(44). Animal studies have yielded mixed results with some Einding the greatest changes in the

vertebrae (2,8) and others finding the greatest change in the femur (45). These differences may

reflect postural differences in the models, since there is less mechanical loading on the spine in a

quadruped. Nevertheless, we believed the most profound changes would be seen in the lumbar

vertebrae because this site has more cancellous bone.

Significance of the Study

This study lays the groundwork for future human studies. It is the first step in a research

sequence that will help define new treatment options to improve the quality of life for patients

exposed to long-term GC therapy and those facing transplant surgeries. This study used

microCT and histomorphometry to compare changes in bone quantity and microarchitecture with

teriparatide treatment in an animal model of GC exposure. This study also allowed further

examination of the specific character of bone loss caused by GC.

Every year nearly 342,000 people in the United States die from lung diseases, which are

the 3rd leading cause of death. Many of these patients are given GC drugs to combat the

inflammatory conditions of their diseases. Patients suffering from cystic fibrosis, emphysema,

and asthma remain on GC treatment for years. A study commissioned by the American Lung

Association estimates that in the United States there are 8.6 million patients suffering from









chronic bronchitis and 3.1 million with emphysema. That study also found that 7.7% of adults

and 8.8% of children and adolescents under age 18 suffer from asthma (46). With better medical

treatments, these patients are surviving longer and it is imperative we find a more effective way

to treat the adverse effects of GCs on bone. Additionally, many end stage lung failure patients

will be evaluated for possible lung transplant procedures. Some transplant centers view

established osteoporosis as a contraindication to lung transplant surgery since patients will need

to take a cocktail of immunosuppressant drugs after transplantation and these drugs have

deleterious effects on bone. This makes finding an effective treatment for GIO imperative. In

the U.S. there have been over 360,000 solid organ transplants since 1988. In 2005, 23,506 solid

organ transplants were performed and 90,620 patients remained on waiting lists (47). Solid

organ transplantation surgery is increasing over time. Advances in medical science have

significantly increased survival times for these patients and preventing or reversing osteoporosis

is becoming a more important quality of life issue.









CHAPTER 2
MATERIALS AND METHODS

Background

This study was designed as a randomized, prospective investigation of the effects of

teriparatide in treating GC-induced osteopenia in mice. There is strong evidence that

glucocorticoids have a deleterious effect on bone. Teriparatide is the only FDA-approved

anabolic bone treatment currently available, but it has not been routinely used to reverse GC-

induced bone loss. This study uses an animal model to assess the effects of the synthetic

glucocorticoid methylprednisolone succinate (prednisolone) and teriparatide at the tissue and

cellular level. To evaluate the efficacy of teriparatide to prevent or reverse GIO, 70 mice were

randomized among 7 treatment groups receiving a combination of prednisolone, teriparatide, or

vehicle to determine changes in bone structure and turnover at the end of a 4-week or 8-week

treatment regimen. At the end of the treatment regimen, bone samples were collected to

determine changes in bone structure and architecture. This protocol was reviewed and approved

by the Institutional Animal Care and Use Committee (IACUC) at the University of Florida.

Animals

The study cohort consisted of 70 male, 7-month old, retired breeder Swiss Webster mice

(Harlan Sprague Dawley, Indianapolis, INT). The Swiss Webster, an outbred mouse strain, was

selected because they are known to have high levels of trabecular bone (7). Swiss Webster mice

achieve peak bone density and cease longitudinal growth in the long bones between 5 and 6

months of age (5) prompting our use of 7-month old mice. Animals had at least seven days to

acclimatize to minimize the effects of stress during shipment. Male animals were used to avoid

the confounding effects of changes in estrogen levels through the lifecycle of female animals.









We used retired breeders because the costs of feeding and maintaining animals until they reached

the age necessary for the study would have been cost prohibitive.

Animal Housing Conditions

Animals were housed in the Special Pathogen-Free (SPF) facility of the Animal Care

Services (ACS) department at the University of Florida. Animals were housed one per cage in

micro-isolator cages that provided animals with filtered air. Cages were kept in a specially

designed rack (American Caging Equipment, Allentown, New Jersey) containing spaces for 70

animals (10 rows with 7 cages per row). Animal cages were moved each week so that animals

rotated to different positions within the rack.

Animals were maintained under standard care conditions with 12 hours light/12 hours dark

in a climate controlled room with an average temperature of 21 degrees centigrade and humidity

of 40%. Animals were fed a standard rodent chow, Teklad Irradiated LM485 mouse/rat chow

(Harlan, Indianapolis, INT). This food was irradiated with Cobalt-60 to kill any bacteria or

viruses present. The rodent chow had a minimum of 19% crude protein and 5% crude fat and a

maximum of 5% fiber. The food contained 0.98% calcium and 0.66% phosphorus. Food intake

was not controlled, but food was weighed every three days to determine each animal's

consumption. Food debris on the cage floor was not measured and was assumed to be

comparable between cages. Water was available ad libitum and was supplied through an

automated watering system with the water purified by reverse osmosis. Animals were also

provided with a supplemental water bottle. All procedures on animals were conducted under a

hood to avoid exposing animals to contaminants. Animals were weighed using a digital balance

at the end of their acclimation period, weekly, and prior to sacrifice. Body weights were used to

monitor animal health.










Study Group Assignment

Study groups, treatments, and timelines are shown in Figure 2-1.

START Weeks 1-4 Weeks 5-8 GROUP

II (Baseline Sacrifice) BSL CNTL (n=10)

I VEH | VEH | 8 week Vehicle Cntl (n=10)

I GC I SACRIFICE 4 week GC Cntl (n=10)

I GC | GC | week GC Cntl (n=10)

I GC | VEH | 4 week GC/Natural Recovery (n=10)

I GC + PTH | GC +PTH | 8 week GC +PTH(n=10)

I GC I GC +PTH | 4 week GC/ week GC+ PTH (n=10)
Figure 2-1. Study Groups, Treatments and Timelines.

Animals were the same age (7 months) at arrival and were block randomized to groups

based on their arrival date at the ACS facility and their body weight. Each week for seven

consecutive weeks, 10 animals arrived and were distributed among treatment groups with 1-2

mice/shipment/group.

Pharmacological Agents

Study Drugs

In this study, prednisolone was used to model glucocorticoid treatment. After 4 weeks of

exposure to prednisolone, the drug was discontinued in one group of animals to assess natural

recovery of the bones. Other animals continued with glucocorticoid treatment alone or were

simultaneously treated with teriparatide to see if that drug could prevent or reverse the effects of

GC on the bones. Animals receiving teriparatide received subcutaneous inj section of 40ug/kg

body weight/day, a dose commonly used in studies using mice (8,35). All study drugs and

vehicles were formulated so that each animal received a subcutaneous inj section volume of

approximately 0.1 ml/injection. Animals were restrained by hand and each animal received two










injections per day. Study drugs or vehicle was administered sequentially at approximately the

same time of day throughout the study. All study drugs were prepared under sterile conditions.

Prednisolone Succinate

Prednisolone (Webster Veterinary Supply, Sterling MA) or vehicle (sterile saline) was

administered at a dose of 2.1 mg/kg/day, 6 days/week. The drug was purchased in liquid form at

a concentration of 20 mg/ml and diluted with sterile saline. Prednisolone was prepared fresh 1-2

times per week under sterile conditions in a biochemistry lab in the College of Health and

Human Performance.

Teriparatide

Teriparatide (Bachem, Torrence, CA) is a recombinant PTH that consists of the same first

34 amino acids found in endogenous PTH. The anabolic action of the drug has been found to

reside in this fragment (48). This drug is currently FDA-approved for use in humans to treat

severe osteoporosis.

Teriparatide was purchased in powder form and dissolved in an acidified, 2% heat-

inactivated mouse serum stock solution (vehicle) using a formulation used in previous rodent

studies (49). Specifically, heat-inactivated mouse serum (obtained from adult, male Swiss

Webster mice) was used to make a stock serum solution that was used as vehicle and to dissolve

teriparatide. The serum stock solution was prepared by mixing 0.1 ml 0.001N HCL and 97.9 ml

sterile saline. The sterile saline and HCL was filtered using a 0.2 micron millipore filter and 2

ml of heat-inactivated serum was then added. The serum stock was divided into 1.5 ml aliquots

and stored at -20 degrees C until needed to dilute the PTH stock solution or for use as vehicle.

To prepare the PTH stock, 1 mg of teriparatide was diluted in 1 ml of the serum stock

solution. The PTH stock solution was then divided into 10 Cl1 and 20 Cl1 aliquots and stored at -

800 C until needed. Storing the PTH stock in small quantities ensured no aliquot was thawed









more than twice. Dissolved teriparatide, was administered at a dose of 40ug/kg/day, 6

days/week subcutaneously in a volume of approximately 0. 1 ml/mouse depending on body

weight.

Flurochrome markers

Flurochrome markers (demeclocycline and calcein) were inj ected to label actively

mineralizing bone. Animals were injected subcutaneously on a pre-determined schedule prior to

sacrifice. These flurochromes bind to calcium and are incorporated into newly forming bone.

They provided a means of determining the amount of bone mineralized between flurochrome

treatments. This is a technique commonly used in histomorphometric analysis of bone formation.

Demeclocycline

Animals were inj ected with demeclocycline (Sigma, St Louis, MO) at a dosage of 15

mg/kg SC 11 and 10 days prior to sacrifice. This drug produces a dull orange fluorescent band

on bone that can be seen under ultraviolet light when it binds to calcium in newly formed bone.

Demeclocycline was purchased in powder form and dissolved in sterile saline. The

mixture was stirred for at least two hours to insure the demeclocycline completely dissolved.

Demeclocycline was prepared fresh the morning of the 11Ith day prior to sacrifice and used for

the -11 and -10 day injections. Any remaining volume was then discarded.

Calcein

Animals were inj ected with calcein (Sigma, St Louis, MO) SC at a dosage of 15 mg/kg at 4

and 3 days prior to sacrifice. This drug binds to calcium in newly forming bone and appears as a

bright green band that can be seen under ultraviolet light. Calcein was purchased in powder

form and prepared for inj section by dissolving it in sterile saline buffered with sodium

bicarbonate. Calcein was prepared fresh the morning of the 4th day prior to sacrifice and used

for the 4- and 3-day injections. Remaining calcein was then discarded.









Anesthesia and Euthanasia

Animals were anesthetized using inhaled isoflurane (2-3.5%) with oxygen as the carrier

gas using an anesthesia cart with a charcoal filter scavenger attached. Animals were placed one

at a time into the anesthesia chamber. The isoflurane gas was started and the animals were

observed until unconscious. They were then removed from the chamber and deep anesthesia

confirmed by a lack of motor responses to a pinch of the foot. The animals were euthanized by

exsanguination from the aorta, followed by cervical dislocation.

Bone Harvesting

Femurs and lumbar vertebrae were harvested to assess (via histomorphometry and

microCT) the bone response to treatment in the appendicular and axial skeleton, respectively.

Both femurs and vertebrae (13th thoracic through 5th lumbar, T13-L5) were excised from each

animal. The femur was disarticulated from the acetabulum and the tibia using a scalpel. A

scalpel was also used to shave off the cranial surface of the distal femur to expose the growth

plate and metaphysis. The bone was then cut at about the mid-point using a hand-held saw or

bone shears.

The lumbar vertebrae were harvested through an incision on the ventral side of the animal.

Internal organs were removed and the ventral portion of the vertebral area gently scraped with a

scalpel to allow visualization of the vertebrae and intervertebral disks. The lumbar vertebrae

were identified by first locating the floating ribs (T11 T13). A cut was made through the

intervertebral disk cranial to the last thoracic vertebrae (T13). The area of the vertebral column

to be removed was identified by counting intervertebral disks, which were visible as white bands

on the ventral aspect of the spinal column. A second incision was made through the

intervertebral disk caudal to the fifth lumbar vertebra. This allowed T13 and lumbar vertebrae

L1-L5 to be removed as one section. Removal of T13 with the lumbar vertebrae made it easier









to identify the cranial and caudal ends of the lumbar vertebrae and, therefore, to identify

individual vertebrae.

The femurs and lumbar vertebrae were stored in 20-ml glass scintillation vials in

phosphate-buffered formalin for 24 hours. After 24 hours, the formalin was poured off and

replaced with 70% alcohol. The bones were then kept at 4 degrees C until microCT and

histomorphometric assessment.

Study Measures

Anthropomorphic Measures

Animal weights were obtained using a digital scale (Ohaus Scout Pro, Pine Brook, New

Jersey). Animals were placed in a weighing bucket to minimize movement and improve

weighing accuracy. Animals were weighed after their acclimation period, weekly during the

study, and prior to sacrifice. Animal weights were used to determine individual drug dosages

and to monitor the health of the animals.

The left femur from each animal was measured using an electronic digital caliper (Little

Machine Shop, Pasadena, CA) to confirm lack of longitudinal growth of the femur over time

among groups. Each bone was measured twice and the average of these measurements was used.

The electronic caliper was zeroed between each measurement.

Histomorphometry

Bone specimen preparation for histomorphometric analysis was carried out at the Wronski

Lab, Department of Physiological Sciences, University of Florida using established protocols

described elsewhere (50,51). In brief, the right femur and lumbar vertebrae L3 were dehydrated

in increasing concentrations of ethanol over a 1-week period and cleared in xylene for 24 hours.

The samples were then embedded undecalcified in modified methylmethacrylate to facilitate

sectioning. The embedding process involved treating the bones in a series of four










methylmethacrylate solutions (with increasing amounts of a catalyst) that progressively

infiltrated the bone over a period of 9 days. The bones were then placed uncapped in a vacuum

dessicator for 6-8 hours. Subsequently, the bones were positioned in the center of the vial to

optimize the sectioning process and placed in a water bath at 420 C over night. The heat caused

the methylmethacrylate to polymerize and harden. Once the methylmethacrylate hardened, the

glass vial was broken and removed leaving a plasticized block containing the bone.

The embedded bones were sectioned longitudinally at 4- and 8 pum thickness using

Leica/Jung 2050 or 2165 microtomes. Six non-consecutive 4 Clm sections and 6 non-consecutive

8 pum sections were cut and mounted on gelatinized glass slides. The two best 4 pum and two best

8Clm sections were selected for analysis. The 4 pum sections were stained for assessment of static

(structural and cell) measurements while the 8 pum sections were coverslipped unstained for

evaluation of dynamic measurements.

Static bone measurements

The 4 Clm thick sections were stained according to the Von Kossa method with a

tetrachrome counterstain (Polysciences Inc., Warrington, PA) (50). This stain causes

mineralized bone to appear black and bone cells and osteoid to stain blue.

Static structural and cellular endpoints were measured in two 4 Clm stained sections using

the Trabecular Analysis System (TAS)/Osteomeasure System (Osteometrics Inc., Atlanta, GA)

or the Bioquant Elite Bone Morphometry System (R&M Biometrics, Nashville, TN). Endpoints

measured or calculated included

* Cancellous bone volume/total volume (BV/TV, %) (percentage of total marrow area
occupied by cancellous bone)

* Trabecular width (Tb.Wi, Clm) (1.99 x B ar/2/ b Pm)

* Trabecular number (Tb.N, #/mm) (BV/TV)/Tb.Th)










* Trabecular separation (Tb.Sp,Clm) ((1/Tb.N) Tb.Th)

* Osteoblast surface/bone surface (Ob.S/B S, %) (percent of bone surface lined by osteoblasts)

* Osteoclast surface/bone surface (Oc. S/B S, %) (percent of bone surface lined by osteoclasts)

For evaluation of structural endpoints using TAS, the bone section was magnified 2x and a

video capture system was used to take an image of the bone. The region of interest (ROI) is 1.5

mm2, beginning 0.5mm proximal to the growth plate and extending back toward the diaphysis.

The ROI was also 0.25 mm from the cortical bone on either side of the femur. TAS software

allowed the user to modify the video image to match the bone section and the software then

calculated the amount of bone present within a region of interest. This data was then used to

calculate Tb.N, Tb.Th, and Tb.Sp as defined above.

Ob.S and Oc.S were measured using the Bioquant Elite Bone Morphometry System. This

system allowed us to manually trace the total perimeter of cancellous bone as well as the portions

of cancellous bone surface covered by osteoblasts and osteoclasts to calculate the proportion of

the total cancellous bone covered by these cells.

Dynamic bone measurements

Dynamic bone analysis was accomplished using flurochrome-based data collected from

unstained 8 Clm femur sections using the Osteomeasure System. Two sections from each animal

were used and the results averaged. Flurochrome data was used to determine

* Mineralizing surface (MS/BS, %) (percentage of cancellous bone surface with a double
flurochrome label; MS/BS is a dynamic index of bone formation).

* Mineral apposition rate (MAR, Clm/day) (distance between the two flurochrome labels
divided by the number of days between label administration; MAR is an index of osteoblast
activity) .

* Bone formation rate/bone surface (BFR/BS, um3/um2/day) (MS x MAR; volume of new
bone formed per unit of total bone surface per unit time)









The slides were magnified at 200X on the microscope and displayed at 250X on the

computer monitor. The area of interest was defined as the area beginning approximately 0.5 mm

proximal to the end of the growth plate and consists of a series of fields that, combined, equal a

cancellous bone area approximately 1.5 mm X 1.5 mm that is about 375 Clm from the cortex.

The cancellous bone (with and without flurochrome labels) was outlined using a digitizing tablet.

Then, the inner and outer flurochrome labels were outlined where double labeling exists and the

distance between these two lines was measured at 4 approximately equidistant points. The

software then calculated the MS, MAR, and BFR/BS as defined above.

MicroCT

MicroCT was used for nondestructive three-dimensional evaluation of bone

microarchitecture. MicroCT identified subtle changes in three-dimensional bone architecture

that cannot be detected by histomorphometry. The bones were scanned using a Scano

microCT40 scanner (Scanco Medical AG, Basserdorf, Switzerland). Cancellous bone in the LV

and femoral metaphysis was evaluated.

Femur

Prior to placement in the microCT, the femurs were first cleaned of non-skeletal

connective tissue and muscle and placed between two thin Styrofoam pads. The Styrofoam pads

help kept the samples from moving during the test. The bones and padding were placed in a

specially designed tube that was 12.3 mm in diameter. Three femurs were loaded into the tube

and scanned sequentially. The tube was then filled with 70% ETOH and covered with parafilm.

The samples were scanned at medium resolution at a voxel size of 12.3 x 12.3 x 12.3 Clm.

Scanning took approximately one hour per bone. Reconstruction of the bones following the scan

took an additional hour per bone. The volume of interest in the distal femur consisted of 1.8

mm2 Starting at the growth plate and moving toward the diaphysis. Of the 120 180 slices










scanned, 150 were analyzed. Twenty slices (approximately 0.25 mm) of cortical bone were also

analyzed in the femoral midshaft. The VOI for cortical bone began at the midpoint of the femur

and included 55 slices toward the proximal femur. Of the 55 slices scanned, 20 were analyzed.

Direct cancellous bone measurements in the femur included:

* total tissue volume (combined volume of cancellous bone and bone marrow in the volume
of interest (VOI)

* Cancellous bone volume (volume in the VOI occupied by cancellous bone)

* Trabecular thickness

* Trabecular number

* Trabecular separation

* Cortical thickness was measured in a sample from the mid-shaft


Once the bone was scanned and reconstructed, drawing tools provided with the software were

used to outline the area of interest (AOI). Every tenth slice was contoured by hand and the

software extrapolated the AOI to the remaining slices. A visual inspection of each slice was

done to ensure no cortical bone was included in the cancellous ROI. The ROI for the distal

femur scanned consisted of the cancellous bone proximal to the growth plate extending to about

1.8 mm toward the diaphysis The cortical bone analyzed began at approximately the center of

the diaphysis to a point 20 uCT slices (250 Cpm) toward the distal metaphysis.

Vertebrae

The second lumbar vertebrae (L2) was analyzed using uCT. First, L2 was separated from

the rest of the vertebrae and non-skeletal connective and muscle tissue was removed. The

vertebrae were scanned eight at a time. The spinal canal of each vertebra was threaded onto a

slender wooden holder and a small piece of Styrofoam was placed at the bottom of the wooden

holder, between each vertebrae, and at the top of the holder. This held the vertebrae upright and










helped controlled movement during the scan. A scout scan was run to identify the area for

analysis and then the scan proceeded automatically. The samples were scanned at medium

resolution at a voxel size of 12.3 x 12.3 x 12.3 pm. The scan took approximately one hour per

bone and required an additional hour per bone reconstruction time prior to analysis. As with the

femoral metaphysis, cancellous bone measurements in the vertebrae included

* Total tissue volume
* Cancellous bone volume
* Trabecular thickness
* Trabecular number
* Trabecular separation.


The AOI for the vertebra included all of the secondary cancellous bone between the two

growth plates. Once the bone was scanned and reconstructed, the software drawing tools

provided were used to contour the AOI at every sixth slice. The software then extrapolated the

AOI to the other slices. All slices were reviewed to ensure no cortical bone was included in the

AOL.

Statistical Analysis

A power analysis was conducted to determine adequate sample size using data from a

study evaluating the effects of treating Swiss Webster mice with the same prednisolone dosage

proposed for this study (5). The present study was designed to achieve a power of at least 0.80.

Power analysis using SAS version 4.0 (SAS Institute Inc., Cary NC) indicated eight animals per

group would result in a 0.87 power. Each group in this study contained 10 animals to insure

study power was not compromised if any animals had to be removed from the study prematurely.

Data is presented in table format as mean a standard deviation (SD) and bar graphs with mean a

standard error for continuous variables, and as percent change for statistically significant

differences between groups. Statistical analysis on data was conducted using SPSS 10.0









statistical software (SPSS Inc, Chicago, IL). Data was analyzed using the nonparametric

Kruskal-Wallis test (52). When significant treatment differences were observed, between-group

comparisons were performed using the Mann Whitney test of independent samples.









CHAPTER 3
LITERATURE REVIEW

This chapter is divided into two parts. Part A will present a general overview of bone

structure and metabolism, including a look at the various types of bone cells that contribute to

the remodeling process. The four phases of bone remodeling: activation, resorption, reversal,

and formation are also described. In the remodeling process, bone is first resorbed and then new

bone is deposited. There is still much we do not understand about this process and how lifestyle,

health, and pharmacologic interventions can influence the balance. The second part of this

chapter describes the effects of two pharmacological agents, one catabolic (prednisolone) and

one anabolic (teriparatide), on bone remodeling. Normal levels of remodeling are altered by

both of these drugs, but in different ways, and Part B of this chapter will describe what we know

about the influences of these drugs on the mouse model.

Bone Biology

Structure of Bone

There are two basic types: cortical and cancellous. Cortical bone is the hard outer layer of

bone and is denser than cancellous bone, which is found closer to the bone marrow. By volume,

cortical bone makes up about 80% of the adult human skeleton. The remaining 20% is the more

changeable cancellous bone. In contrast to the hard and only slightly porous cortical bone,

cancellous bone is a complex three-dimensional network of curved plates and rods in close

association with bone marrow and is enclosed by cortical bone. Cancellous bone is made up of a

lattice of large plates and rods collectively called trabeculae. The inner or endocortical side of

cortical bone within the medullary cavity (53).

Cancellous bone is mainly found in bones of the axial skeleton, in flat and irregular bones,

and in the ends of the long bones (53). This type of bone experiences deformation when loaded










and is better able to bear loads without becoming damaged (54). The lattice-like structure of

cancellous bone means its surface-to-volume ratio is higher than that of cortical bone. Since

remodeling takes place at the surface of bones, the greater relative surface area of cancellous

bone means remodeling takes place there at a rate ten times greater than in cortical bone (53,55).

Thus, when there is an imbalance leading to more bone resorption than formation, the effects

will be most apparent in cancellous bone such as that found in the vertebrae and the ends of long

bones.

Bones do not contain cortical and cancellous bone tissue uniformly. A typical long bone

has 3 regions that vary in composition. The diaphysis, the shaft of the bone, is comprised mainly

of cortical bone. The ends of the long bones, known as the epiphyses, are to a large degree

cancellous bone, as are the metaphyses, the conical section of bone connecting the epiphysis and

diaphysis. Cancellous bone is made up of plates of bone tissue and loss results in a gradual shift

from plate-like to rod-like structures as the dominant elements (56). This contributes to

increased fragility of the bone, since spaces within the bone increase as the "struts" connecting

one section to another disappear. Electron microscopy of an older person's bones shows wider

spaces and fewer structural connections. This has important implications for bone strength since

trabecular struts, once lost, cannot be replaced (56). Trabecular compromise occurs when

osteoclasts erode a cavity too deeply or when the osteoblasts are unable to lay down a sufficient

amount of replacement bone (56).

The maj or components of cortical and cancellous bone are type 1 collagen, water,

hydroxyapatite mineral, and small amounts of proteoglycans and noncollagenous proteins. Type

1 collagen is a structural protein found mainly in bone and tendons. Hydroxyapatite,

Calo(PO4)6(OH)2, makes up virtually all of the mineral in bone and represents the maj or










storehouse for the body's calcium. Calcium is taken in or released based on fluctuating plasma

calcium levels and the presence of the maj or calcium regulating hormones, PTH and calcitonin.

Bone Cells

Bone cells include osteoclasts, osteoblasts, osteocytes, and bone-lining cells. Each cell

type is critical to bone remodeling and these cells have complex mechanisms of communication

that control their actions and interactions. These cells generally operate in balance with each

other, although aging, some disease conditions, and certain drugs can alter that balance. There is

ample opportunity for the balance in bone metabolism to shift toward more resorption than

deposition since the osteoclast is able to resorb in one day an amount of bone that osteoblasts

need several days to replace.

Osteoclasts

Osteoclasts are large, multinucleated cells associated with bone resorption. Osteoclasts

originate from hematopoietic stem cells in the bone marrow and travel via the circulatory (or

perhaps the lymphatic) system. Mature osteoclasts are responsible for bone resorption, where

bone is broken down and the calcium within liberated. Osteoclasts adhere to bone by means of

an actin ring that is anchored to the extracellular matrix by integrins. This forms a sealing zone

that creates a microenvironment between the osteoclast and the surface of the bone that will be

resorbed (57). When osteoclasts attach to bone, the cell is polarized and generates a ruffled

border. It is at this ruffled border that vesicles containing cathepsin K and membrane-bound H+

ATPase exist. Cathepsin K, an acidic collagenase, degrades the organic component in bone

(type 1 collagen) while the H+ ATPase secretes hydrochloric acid into the sub-cellular space and

dissolves hydroxyapatite (58).

Osteoclasts are found in cavities on the bone surface, which they themselves form, called

resorption pits or Howship's lacunae. Interestingly, although associated with resorption,









osteoclasts have no receptors for PTH, the main endogenous mediator of bone breakdown.

Instead, osteoclasts have receptors for calcitonin, a hormone that inhibits bone resorption (59).

Osteoblasts

Mature osteoblasts are bone-forming cells that typically reside on the bone surface where

they secrete unmineralized matrix, called osteoid, during the bone formation process. They also

participate in calcification of bone and regulate the movement of calcium and phosphate into and

out of the bone. These cells are normally cuboidal in shape. Osteoblasts themselves produce

and secrete a number of substances important to bone metabolism including type 1 collagen,

non-collagenous matrix proteins such as osteocalcin and osteonectin, growth factors,

prostaglandins El and E2, Receptor Activator of Nuclear factor kappa B ligand (RANKL) and

osteoprotegerin (OPG) and cytokines including interleukin (1L)-1, IL-6, and 11, TNF, and TGF-P

(60). Although osteoblasts are most noted for bone formation, they also help control bone

resorption since they have receptors for PTH and secrete OPG and RANKL.

In humans, osteoblasts secrete osteoid at the rate of about 1 micrometer per day. This

means that either the lifespan of the osteoblast is quite long or that multiple generations of

osteoblasts are involved in refilling a given resorption pit since these pits can be quite deep

(61,62). When the osteoblast has finished secreting osteoid, it returns the preosteoblast pool,

transforms into a bone-lining cell, gets buried as an osteocyte, or dies (53).

Osteocytes

Osteoblasts which become trapped in the osteoid they secrete are called osteocytes. Each

lacunae contains only a single osteocyte. These cells maintain contact with each other and with

bone-lining cells via slender processes that reach through the canaliculi of the bone at the gap

junctions. There are gap junctions between adj acent bone-lining cells and between bone-lining

cells and osteocytes. Osteocytes are thought to be involved in detecting microfractures and the









cell signaling that begins the process of remodeling (53,61,62). They may also be involved in

storing mineral ions following a meal rich in calcium and in transporting minerals from deeper

skeletal reservoirs to the extracellular fluid compartment after resorption (53).

Bone Lining Cells

The final major bone cell type is the bone-lining cell. These are long, flat cells that cover

quiescent (or resting) bone surfaces, where bone is neither being resorbed nor formed. Like

osteocytes, bone lining cells originate from osteoblasts. They differ from osteocytes, however,

in that they remain on the bone surface rather than being buried in the matrix. As bone

formation ends, bone lining cells remain on the newly formed bone surface. They communicate

with osteocytes and each other through gap junctions. The bone lining cells assist the osteocytes

in moving mineral in and out of the bone and may also play a role in sensing mechanical strain

on bone (63).

Bone Remodeling

Bone remodeling is the term used to describe the processes of resorbing old bone and

depositing new bone at the same site. Bone remodeling is an on-going "housekeeping" activity

of healthy bone. Even in the absence of external stimuli there will be remodeling activity. This

process of remodeling is most obvious in its accelerated form when there is a fracture and the

body quickly moves to repair the damage. On a more subtle scale, however, the body is

constantly replacing old bone and repairing microfractures, the damage caused by daily activity.

Repair of this damage helps keep the bones strong by preventing structural weaknesses from

accumulating. Continual remodeling helps bone maintain strength and structural integrity, so

long as there is a balance between resorption and deposition. In the aggregate, what determines

whether more bone is being formed or resorbed is the relative amount of each activity.

Hormones such as estrogen, growth hormone, insulin, parathyroid hormone (PTH), testosterone









and agents like fluoride and aluminum directly or indirectly affect the balance to varying degrees

(53,61,62,64,65).

Basic Multicellular Units (BMUs) orchestrate bone turnover, removing mechanically

unneeded bone and repairing microdamage by laying down new bone. BMUs consist of

osteoclasts and osteoblasts that congregate at a specific area of the bone where they have been

drawn, possibly by the signaling activity of osteocytes (66). In cortical bone the BMU tunnel in

a cone-like pattern through the bone while in the trabecular bone BMUs scallop the surface of

the trabeculae to form a trench (39,43). The BMU remodels bone in four distinct phases:

activation, resorption, reversal, and formation. When not involved in remodeling, bone is said to

be quiescent.

Bone remodeling begins when a quiescent skeletal surface is activated. The activation

phase is characterized by a retraction of the bone lining cells at the activation site and formation

of new blood vessels that will bring osteoclasts to the resorption site. This exposes mineralized

bone surface which may act as a chemoattractant for osteoclast precursor cells (53,61,62).

During the activation phase, preosteoclasts fuse to form the characteristic multinucleated mature

osteoclasts which will attach themselves to the exposed bone surface (43).

The resorption phase begins when the multinucleated osteoclasts begin to resorb bone at

the remodeling site. The osteoclast, with its ruffled border, attaches to the area of the bone to be

resorbed; here the osteoclast and the underlying bone form the microenvironment into which the

osteoclast secretes acids. When the osteoclast secretes acids into this microenvironment, the

collagen matrix breaks down, forming concave pits called resorption pits or Howship's lacunae.

In humans, the resorption pits have an average erosion depth of 60 micrometers in trabecular

bone and about 100 micrometers in cortical bone. The osteoclasts can erode up to tens of










micrometers per day (61). The whole process of resorption takes 1-3 weeks and culminates with

the release of calcium and other compounds from the matrix of the dissolved bone. The end of

resorption is marked by osteoclasts migrating from the bone surface to nearby marrow spaces,

where they hibernate or die (67).

The third stage of remodeling, reversal, is characterized by preparations to lay down new

bone. Phagocytes smooth out ragged edges left by the osteoclasts and a thin layer of collagen

and matrix, referred to as a cement line, is laid down. Osteoblasts are drawn to the area through

as yet not understood mechanisms. Osteoblasts may be stimulated to mature by signals from

compounds such as growth factors released from the bone itself when it breaks down. Some

believe these signals occur when the calcium released by resorption activates calcium receptors

on osteoblasts. This, theoretically, helps insure bone resorption does not get out of control, since

the more resorption there is, the more osteoblasts would be stimulated to form new bone (68).

Once at the remodeling site, osteoblasts adhere to the cement line where they begin filling in the

resorption cavity with osteoid (43). In the final phase of remodeling, known as formation, new

osteoid is secreted by the osteoblasts. Under normal conditions, if there is sufficient calcium

available, the new bone is mineralized. Flurochrome labeling is often used to measure this

process.

In humans, the entire sequence of resorption and formation at a given remodeling site takes

place over a period of several months and it is estimated that the lifespan of a BMU is about 6-9

months (43,64,68). In healthy adults between 3 and 4 million BMUs are activated annually,

with about 1 million operating at any given time (43). The remodeling cycle in animals follows

the same sequence of events but the time required is significantly less. This accelerated response









in animals makes them an attractive model to predict the effects of conditions and treatments in

humans

Remodeling Balance : The RANKL~/OPG/RANK Axis

The balance of bone remodeling is controlled by hormones and paracrine influences

originating from osteoblasts or stromal cells. The discovery of RANKL and OPG was an

important milestone in bone research since this helped explain a seeming paradox of bone

remodeling. Namely, that many of the hormones, cytokines, and growth factors that regulate

osteoclast activity have receptors on the osteoblast (69). Researchers had also noted that cell

cultures of osteoclast precursors physically separated from osteoblasts did not develop into

functional osteoclasts (65) and osteoclast apoptosis increased, indicating a close relationship

between osteoblasts and the production and differentiation of osteoclasts. RANKL is produced

by osteoblasts and binds to RANK receptors on osteoclasts and osteoclast precursors where it

stimulates differentiation and greater activity of osteoclasts. OPG is also produced by

osteoblasts but is a decoy receptor that competitively binds RANKL. The amount of bone

resorption is modulated at least to some extent by the ratio of RANKL to OPG.

It is the ratio of RANKL to OPG, and not just the level of RANKL, that seems to govern

whether bone remodeling favors formation or resorption so it is important to see what substances

influence the ratio (70). OPG production is stimulated by 1,25 dihydroxy vitamin D3, BMP-2,

TNF-a IL-la and -10, and estrogen (1,71). The resulting increase in OPG production removes

additional amounts of RANKL, thus decreasing the amount of RANKL available to bind to

RANK which results in reduced osteoclast differentiation and activity. Many circumstances lead

to increased RANKL production, such as glucocorticoid use, lack of estrogen, and it is often

present in diseases like rheumatoid arthritis.









Effects of Glucocorticoid Drugs and Teriparatide on Bone

GCs inhibit the formation of osteoblasts and osteoclasts, increase apoptosis in osteoblasts,

and interfere with normal bone remodeling. While the negative effects of GCs on bone have

long been recognized, the mechanisms remain to be fully understood (72-75). Glucocorticoids

bind to a cytoplasmic glucocorticoid receptor (GR) found on osteoblasts (76). The receptor is a

ligand-operated transcription factor. When not bound, the receptor is located in the cytoplasm as

a protein complex. When activated, the complex dissociates and the receptor moves into the

nucleus and binds to regulatory elements in the promoter regions of certain anti-inflammatory

genes (77-79). The GR also inactivates inflammatory genes by binding to transcription factors

activator protein-1 (AP-1) and nuclear factor kappaB (NF-KB) (77,78) With these transcription

factors bound there is inhibition of pro-inflammatory cytokines such as IL-1P, IL-4, IL-5, and

IL-8, and TNF-a (77). Genomic effects begin probably no sooner than 30 minutes after GC

administration, and are initiated by binding of the steroid to cytosolic receptors. Nongenomic

effects occur sooner, often within a few minutes, and are mediated by membrane-bound GC

receptors (74). There appears to be general agreement that GCs cause decreased bone formation;

the case is not so clear for bone resorption (21,23).

Systemic Effects of Glucocorticoid Drugs

GCs exert a number of effects, both direct and indirect, that influence bone metabolism.

(see Table 3-1). The direct effects of glucocorticoid drugs are those that effect the bone cells

themselves and includes actions that effect osteoclasts, osteoblasts and osteocytes. Direct effects

also include influences on the production of RANKL and OPG and influences on various bone-

related growth factors. Indirect effects include influences on organ systems that influence

calcium metabolism.










Table 3-1. Common Effects of Glucocorticoid Therapy.
Direct Effects Indirect Effects
Increased Osteoclast formation Increased Urinary Calcium Excretion
Increased Osteoblast apoptosis Increased Intestinal Calcium Absorption
Increased Osteocyte apoptosis Decreased GH production
Increased RANKL Hypogonadism
Decreased Osteoblast #s and activity Impaired renal function
Decreased OPG Secondary Hyperparathyroidism
Decreased Type I Collagen Production
Decreased skeletal growth factors
(IGF-1, TGF-P)
RANKL = Receptor activator of nuclear factor (NF)-kB ligand; OPG = Osteoprotegerin; IGF-1 = Insulin-like
Growth Factor; TGF- B = Tumor Growth Factor- B; GH= Growth Hormone.

Direct Effects of Glucocorticoids

The effects of GCs on osteoblasts are potent, causing pre-osteoblasts to differentiate to

adipocytes and decreasing synthesis of type I collagen by mature osteoblasts (80,81). GCs also

decrease the ability of osteoblasts to adhere to the extracellular matrix and promotes matrix

breakdown by stimulating the activity of interstitial collagenase. Additionally, GCs amplify the

response of osteoblasts to endogenous PTH by increasing the number of PTH receptors on the

cell (23). The osteoblast lifespan decreases, leaving less time for synthesis of bone matrix and

mineralization. Taken together, these effects result in a significant decrease in bone formation as

evidenced by sharp reductions in circulating levels of osteocalcin even at low doses (~ 5mg/day

in humans) of GC (76,82). GCs also inhibit a number of growth factors such as IGF-1, which

increase the synthesis of type I collagen, and decrease collagenase 3 expression (80).

Although there is general agreement that GC use decreases bone formation, it is less clear

whether GCs increase bone resorption (23,74,83). Studies of bone cells in vitro have variously

shown stimulation and inhibition of osteoclasts in cell cultures (23,83) in response to GC, and

decreased apoptosis of mature osteoclasts (75). Some histomorphometric studies have found

increased resorption (84) in the presence of GCs but serum and urine markers of bone resorption

have shown inconsistent results (84).









GC treatment increases osteoblast production of RANKL and colony-stimulating factor

(CSF)-1 (also known as macrophage-colony stimulating factor or M-CSF) (81). The

combination of M-CSF and RANKL stimulates osteoclastogenesis. At the same time RANKL is

increasing, OPG levels decrease. GC drugs have been shown to inhibit OPG mRNA by 70-90%,

increase mRNA levels of RANKL and RANKL/M-CSF-induced TRAP activity by over 50%

(1,60,85). This has the potential to shift the bone remodeling balance in favor of resorption by

increasing osteoclast formation.

With GC use in humans, there appears to be an early increase in bone loss which

moderates over time, creating a biphasic response (76,80,86,87). One explanation for early

increases in resorption that subside later is the influence of GCs on induction of IL-6 receptors in

bone (87). Since IL-6 is a cytokine important in osteoclast recruitment, any increase in the

number of receptors in skeletal tissue could increase bone resorption by recruiting more

osteoclasts. At the same time, GCs also inhibit osteoblastogenesis. Declining numbers of

osteoblasts eventually will produce less aggregate RANKL, causing reduced osteoclastogenesis

as well (88). That may explain observations that the greatest bone losses from GC use are

experienced early in treatment and the rate of loss decreases and levels off over time.

Indirect Effects of Glucocorticoid Drugs

In addition to the direct effects of GC on bone metabolism there are also indirect effects

that similarly result in bone loss over time. The indirect effects of GCs on bone involve a

number of organ systems in the body and are summarized in Table 3-1.

Decreased intestinal absorption of calcium

GC use results in a decrease in calcium absorption in the intestines (80). Although the

mechanism is not entirely understood, GCs appear to effect the duodenum by inhibiting active









calcium transport, decreasing the production of calcium-binding proteins and possibly increasing

the degradation of 1,25(OH)2 vitamin D at its binding site (21,74,76,86).

Increased renal elimination of calcium

Increased renal excretion of calcium may be due to a reduction in reabsorption of calcium

in the distal tubule of the kidney (74,76). In the presence of GCs, the kidney tubules handle

sodium and calcium cations differently (86). GC treatment increases activity of epithelial Na

channels, passive sodium channels on the apical membrane of the distal tubules and conducting

duct cells and this increases the activity of Na / Ca2+ antiport pumps, resulting in increased

calcium extrusion (86). Increased renal calcium excretion coupled with decreased intestinal

absorption may lead to secondary hyperparathyroidism (23,74). It is unclear how this affects

overall bone remodeling, however. It was once believed that secondary hyperparathyroidism

accounted for GC-mediated changes in bone, but research now suggests the situation is much

more complex and dynamic (72). Even in cases where secondary hyperparathyroidism occurs, it

does not explain the trabecular bone loss seen with GC use (80). In patients with secondary

hyperparathyroidism, bone remodeling is increased (80) and the main effects are seen in cortical

bone (81) instead of the decreased remodeling that primarily affects cancellous bone as seen in

GIO.

Antagonistic action on gonadal functions

Researchers have identified a direct GC-mediated effect on the production of gonadal

steroids in men and women (74). It is believed GCs suppress the hypothalamic-pituitary-adrenal

axis and inhibit gonadotropin secretion (84). GC treatment has been shown to decrease

circulating levels of testosterone in men by about 50% (75). Similar effects on estrogen are

believed to occur in women (18,75). Both estrogens and androgens suppress bone resorption by









inhibiting osteoblastic release of local stimulating factors that cause formation of increased

numbers of osteoclasts (17,75).

Increased sensitivity to PTH

In vitro studies of isolated bone cells have shown that GCs modulated PTH sensitivity of

both osteoblasts and osteocytes such that lower levels of PTH still elicited measurable

biochemical changes (18). This may be accomplished by GC-mediated upregulation of

osteoblast PTH receptors (74) or increased affinity of the receptor for PTH (23,81). This could

explain why changes in bone are seen even when PTH levels remain in the normal range.

Bone loss in response to glucocorticoid treatment

Studies have shown that with GC treatment, there is a loss in trabecular connectivity

making this population more susceptible to fracture (23,76,81). This change in bone

microarchitecture cannot be detected with densitometry, the most common clinical means of

testing for bone loss. Some have suggested that the GC-mediated loss of osteoblasts and

osteocytes compromise bone strength independent of bone loss (72). According to this theory,

the integrity of bone relies on the network of osteocytes found there. Osteocyte apoptosis may

reduce the signaling available to initiate the replacement of damaged bone (73). This may

explain why fracture risk increases as early as three months after GC use, even before significant

bone loss has occurred (73). In response to this, the Royal College of Physicians of London in

recent years suggested using a T score of -1.5 or lower as the treatment threshold for GC users

(81). Despite these recommendations, it is estimated that less than 15% of those on long-term

steroid treatment also receive preventive medication to prevent osteoporosis (16,22,89) and there

still appears to be limited testing for bone loss in GC patients (22). There have been suggestions

that preventive treatment should begin at the same time as GC therapy is initiated (89).









Parathyroid Hormone

Dual Nature of PTH : Continuous versus Intermittent Administration

Parathyroid hormone has a surprising, dual effect in mammals depending on whether

delivery is continuous or intermittent. At least since the 1930s researchers have noted that PTH

is catabolic when exposure was continuous (as it normally is in the body in response to low

plasma calcium levels) but showed anabolic properties when administered intermittently (90-

93). The use of PTH in an anabolic role was not seriously pursued, however, until it became

possible to manufacture synthetic PTH. In the mid-1970s, recombinant techniques made it

possible to sequence the amino acid fragment responsible for the hormone's anabolic effect (90).

This anabolic capability resides in the first 34 amino acids on the N-terminal end which is why

teriparatide is manufactured as PTH (1-34) (48,92,94).

It has been suggested that at least some of the anabolic effects of intermittent PTH in bone

are mediated through an IGF-I-dependent mechanism, while the catabolic effects are mediated

through gene expression that causes an increased ratio of RANKL to OPG (95,96). Teriparatide

reverses the effects of GC on IGF-I expression in vitro, and this may partially explain its effects

in treating GC-induced bone loss (80).

Parathyroid Hormone (PTH 1-84)

Endogenous PTH is an 84-amino acid protein secreted by the chief cells of the parathyroid

glands when low serum calcium levels are detected by calcium receptors on the parathyroid

glands (97) or there are elevated levels of extracellular phosphate (98). PTH 1-84 is a critical

mediator of calcium homeostasis. Calcium-sensing receptors on the surface cells of the

parathyroid gland respond to minute-by-minute changes in serum calcium levels (48,94,99) and


To avoid confusion, PTH produced by the parathyroid glands will be referred to as endogenous PTH or PTH 1-84
and recombinant parathyroid hormone will be referred to as teriparatide or PTH (1-34).









maintain calcium balance through direct actions on bone and kidneys and indirectly through the

gastrointestinal tract. When blood levels of calcium are low, PTH stimulates bone resorption to

liberate calcium stored in the bone matrix and enhances calcium reabsorption at the distal

convoluted tubules of the kidney (94,98). PTH also regulates 1 alpha-hydroxylase activity in the

kidney, facilitating the conversion of 25-hydroxyvitamin D to 1,25 dihydroxyvitamin D in the

kidney, which then acts on the GI tract to stimulate increased absorption of calcium across the

gut (94,99,100).

In humans, the half-life of endogenous PTH in the blood is less than 3 minutes and it is

metabolized by both the kidney (20-30%) and liver (60-70%) (90,94). This rapid metabolism

means the availability of PTH is determined by the rate of secretion from the parathyroid glands.

Endogenous PTH has both rapid and slow effects on bone. The rapid phase occurs within 30-90

minutes after exposure and is characterized by increased osteoclast activity. A second, later

phase, is associated with an increase in both the number and activity of osteoclasts (94,101).

With continuous PTH exposure there is a decrease in OPG mRNA and an increase in RANKL

mRNA.

While the actual mechanisms are not fully understood, PTH may exert its actions by

activating a number of enzymes such as collagenase, lysosomal hydroxylases, acid phosphatases,

H K -adenosine triphosphatases, Na /Ca' exchange systems, cathepsin B, or cystein proteases.

Continuous PTH exposure causes bone lining cells to retract from the bone surface as part of a

calpain-dependent modification to the osteocyte cytoskeleton. This allows osteoclasts to attach

to the bone surface where they then initiate the process of bone resorption (94). PTH also

inhibits osteoclast apoptosis, possibly by stimulating the expression of RANKL and decreasing

the expression of OPG by osteoblasts (102).









Teriparatide (PTH 1-34)

Teriparatide, a recombinant parathyroid hormone marketed as FORTEOTM by Eli Lilly

pharmaceutical company, is currently the only FDA-approved anabolic drug for osteoporosis.

Unlike the anti-resorptive drugs, teriparatide stimulates bone formation, which contributes to

increased bone mass, quality, and strength. Teriparatide is administered as a subcutaneous

injection typically 20ug/day for 18-24 months in humans. The drug is not recommended for

long-duration use since it caused an increased incidence of osteosarcoma in rats after long-term

exposure to large doses of the drug (103-105). Teriparatide is manufactured from a strain of

Escherichia coli modified by recombinant DNA technology (100,106). Teriparatide, which has a

bioavailability of around 95% (90,100), reaches peak plasma concentration in about 30 minutes,

then drops to virtually undetectable levels in 3-4 hours (90,92,97,100). The systemic clearance

of teriparatide is approximately 62 L/hour in women and 94 L/hour in men, which is greater than

the rate of normal hepatic plasma flow, indicating both hepatic and kidney clearance similar to

PTH (1-84) (100).

Following extensive testing and clinical trials, teriparatide was approved by the FDA in

November 2002 (100) for the treatment of postmenopausal women with osteoporosis who are at

high risk for fracture and to increase bone mass in men with primary or hypogonadal

osteoporosis who are at high risk for fractures (107). Daily inj sections of teriparatide increase

bone mass, microarchitectural structure and bone strength in mice, rats, rabbits, monkeys and

humans (97). Studies have shown it provides a statistically significant increase in BMD at

clinically important sites such as the lumbar spine (29). The increased bone turnover caused by

teriparatide results in additional bone apposition on both periosteal and cancellous bone surfaces

(97). The effects of teriparatide are less robust at the femoral neck.









Mechanisms of Action

PTH and teriparatide work through G Proteins and the PTH Receptor 1 (PTH1R) has a

similar affinity for both (90). PTH receptors are found predominantly on osteoblasts (but are

also found in renal tubular cells) (48,90,94). Binding of PTH to the receptor activates adenylate

cyclase and phospholipases A, C, and D and increases intracellular levels of cAMP and calcium

(90). Some think it may be the ability of PTH to stimulate both adenylate cyclase and

phospholipase C that gives it its dual anabolic and catabolic abilities (48).

The primary effects of teriparatide on osteoclasts are indirect and are mediated by the

drug's effects on osteoblasts. Increased bone formation following exposure to teriparatide

occurs because of an increase in osteoblast numbers either through enhanced differentiation of

pre-osteoblasts or an increase in the number of existing bone lining cells that differentiate into

osteoblasts (48,90, 108). Teriparatide also has an anti-apoptotic effect on osteoblasts, enabling

them to secrete bone matrix for a longer period of time (48,92,100, 109). Through these actions,

the production of collagen-based bone matrix increases, improving trabecular bone volume and

connectivity (90, 110). Teriparatide also acts on the cortical surface of cortical bone and

increases its thickness without increasing porosity (90,109, 111).

Treatment with teriparatide has a variety of effects on osteoblasts, including causing

increased secretion of various growth factors, such as transforming growth factor P (TGF-P), IGF

I and II, IGF binding proteins, bone morphogenic proteins (BMPs), and cytokines such as 1L-1,

Il-6 and M-CSF (48,92,95,112). M-CSF, IL-1, RANKL, and TNF P all enhance osteoclast

survival (102). The net result is increased bone turnover. However, unlike the increased

turnover seen in many disease conditions, the increased turnover associated with teriparatide

favors bone formation over resorption. This is what causes teriparatide to have its anabolic

effect on bone.









Studies of Glucocorticoid-Induced Bone Loss in Mice

Validity of the Mouse as a Model of Glucocorticoid-Induced Bone Loss

The mouse genome has now been sequenced, making this animal an even more attractive

model for scientific research. The mouse has shown its value in studies of bone loss due to aging

and sex steroid alterations (3) and some believe it is the preferred rodent model for the study of

bone loss due to GC exposure (3-5). This is because some researchers believe the rat may be

resistant to the deleterious effects on bone associated with GC exposure and therefore may not

represent a good model for this specific condition (36-3 8). Commonly used mouse strains to

assess the effects of GCs include the Swiss Webster (5), the C57B1 (4) and the Balb/C (3) (see

table 2). Studies to date have used techniques such as histomorphometry, microCT, serum

biochemistry, and DXA to measure bone responses in mice exposed to GCs (2-5). Most studies

have reported that GC treatment induces greater axial than appendicular bone loss (5) without

significant weight loss or hypogonadism (2,3). A summary of GC-induced bone loss studies

using mice listed in Appendix B.

Glucocorticoid-Induced Bone Loss in Mice

Glucocorticoid drugs have been shown to affect mouse bone metabolism during both in

vitro and in vivo experiments. In cell culture studies, the number of osteoclast precursors

following treatment with prednisolone for 4 or 10 days decreased significantly after just 4 days.

Osteoblast precursors also decreased, but only after 10 days of treatment. Despite fewer

precursor cells, prednisolone exposure resulted in an 81% increase in osteoclast numbers and a

20-fold increase in the ratio of osteoclasts to osteoblasts perhaps reflecting decreased osteoclast

apoptosis (4).

In adult Swiss Webster mice, glucocorticoids affect osteoblasts and osteocytes at a

threshold dose of 1.4 mg/kg body weight (5). Some researchers have used this threshold dose (2)









while others have used a dose of 2. 1 mg/kg body weight (4,5). At this higher level of treatment,

researchers have found increased osteoclast survival in as few as 10 days (4). MicroCT analysis

following administration of low levels of GC (1.2 mg/kg) only found significant differences in

BV, while there were no significant differences either in Tb.N or Tb.Th (2). Bone changes were

generally dose and duration dependent. In one study the greatest changes were detected when 10

mg/kg was administered over a 21-day period (3).

Histomorphometric analysis of the bones has yielded mixed results with some studies

finding significant decreases in trabecular thickness while other studies found no significant

difference in this measure (2-5). In these studies, histomorphometric analysis of kinetic

measures were more consistent and showed significant decreases in mineralizing surface,

mineral apposition rate, and bone formation rate (2-5). Studies have also demonstrated a

preferential bone loss in the axial skeleton (5) and that spinal BMD decreases in a dose-

dependent manner (3-5).

Dynamic measures, such as mineralizing surface (MS), mineral apposition rate (MAR),

and bone formation rate (BFR/BS) which are only available using histomorphometry, showed

changes in Swiss Webster mice at dosages > 1.2 mg/kg (2,4,5) while it took dosages of 10mg/kg

to elicit changes in Balb/C mice (3).

Teriparatide Treatment in Mice

Teriparatide use has resulted in increased bone density and strength when used in mice

(2,4,5,7,8). There have been studies treating intact and ovariectomized mice with teriparatide

(see table 3-2), though none of these studies examined the effectiveness of teriparatide in

preventing or reversing GC-induced bone loss in mice. Studies most commonly used the

C57BL/6 (8,35,45), CBA-1 (1 13), and Swiss Webster strains (7) of mice. In the strains of mice









tested, teriparatide had the greatest anabolic effect in the femur and was dose and duration

dependent (7,8,45).

One study found a subcutaneous dose of 40Clg/kg/day 5 days/week increased BMD within

1-2 weeks in the tibia and within 7 weeks in the vertebrae, suggesting site specific differences

occur (45). This finding differs from another study which found the earliest effects in the

vertebrae (8). Researchers have reported increased bone mass in cortical as well as cancellous

bone, although increases in cortical bone were found primarily in long bones, possibly reflecting

a response to mechanical loading patterns (8). Significant effects from teriparatide seem to

depend on the presence of existing trabeculae and may be hampered in areas that have suffered

severe disruption of trabeculae (8). A summary of studies using PTH in mice is in Appendix B

Further Considerations

We do not fully understand the mechanisms governing GC-induced osteoporosis nor how

they effect bone mass and microarchitectural structure. This study seeks to further our

understanding of these processes in mice as a prelude to human studies. To date, there has only

been one study examining the efficacy of teriparatide in treating glucocorticoid-induced

osteoporosis. In this study, postmenopausal women on long-term GC therapy and hormone

replacement therapy (HRT) also received teriparatide. The addition of teriparatide resulted in

significant bone density increases in the axial skeleton (6). No study ofteriparatide alone in

glucocorticoid-induced osteoporosis in a human or murine model has been done to date.










CHAPTER 4
RESULTS

Measurement Design

During this experiment, 70 7-month old male Swiss Webster mice were randomized into 7

groups which received prednisolone, teriparatide, or vehicle in an attempt to characterize the

effects of teriparatide on bone in glucocorticoid-treated mice. Groups are identified in Table 4-1.

At the end of the study, mice were euthanized and both femurs and lumbar vertebrae 2 and 3 (LV

2, LV3) were removed and analyzed using histomorphometry or microCT. Histomorphometric

techniques were used on the right femur and L3 to measure the static parameters of BV/TV,

Tb.N, Tb.Wi, Tb.Sp and dynamic parameters of MS, MAR, and BFR/BS. MicroCT measures

included BV/TV, Tb.N, Tb.Th, and Tb.Sp in the left femur and L2. Cortical BV/TV was also

measured in the mid-shaft of the left femurs. Table values are reported as mean + standard

deviation and figure values are reported as mean + standard error.

Table 4-1. Experimental Groups and Description of Treatments.
Group n = Treatment (6 days/week)
BSL CNTL 10 Baseline Control

8 WK CNTL 10 8-Week Control. Received GC-vehicle and
PTH- vehicle for 8 weeks

GC4/SAC 10 Received GC and PTH-vehicle for 4 weeks and
were then sacrificed

GC4/RECV 10 Received GC and PTH-vehicle for 4 weeks and
then received GC-Vehicle and PTH-Vehicle for
4 weeks to allow for natural recovery

GC8 10 Received GC and PTH-vehicle for 8 weeks

GC4/GC-PTH4 10 Received GC and PTH-vehicle for 4 weeks and
then GC and PTH for 4 weeks

GC-PTH8 10 Received GC and PTH for 8 weeks
GC = prednisolone, 2.1 mg/kg/day; GC-vehicle = sterile saline; PTH = teriparatide 40ug/kg/day; PTH-vehicle = 2%
acidified mouse serum.










Anthropomorphic Measures

Animals were weighed at study entry, weekly, and prior to sacrifice. Weight changes were

used to monitor animal health and adjust dosages of study drugs. Weight data are presented in

Table 4-2. There were no statistical differences in the average weight between the groups either

at the start or end of the experiment.

Table 4-2. Mean Animal Weights by Group.
GROUP Start Wt (g) End Wt (g) Diff (g)

BSL CNTL (n =10) 35.8 + 2.7 n/a

8 WK CNTL (n =10) 35.2 + 3.4 36.1 + 2.6 + 0.9

GC4/SAC (n = 10) 35.3 + 3.7 34.0 + 4.2 1.3

GC4/RECV (n = 10) 35.7 + 2.5 36.9 + 3.5 + 1.2

GC 8 (n = 10) 35.2 + 2.3 34.4 + 3.7 0.8

GC4/GC-PTH4 n = 10) 35.6 + 1.8 35.1 + 2.8 0.5

GC8/PTH8 (n = 10) 35.1 + 4.0 35.1 + 4.2 0.0

Values are expressed as mean + standard deviation.

Previous studies indicate 7-month old Swiss Webster mice have reached skeletal maturity

and have ceased longitudinal bone growth (5). To verify cessation of long bone growth, each

animal's left femur was measured and the results are presented in Table 4-3. There were no

significant differences in femur length between the groups.

Table 4-3. Mean Femur Lengths by Group.
BSL CNT 8 WK GC4/ GC4/ GC 8 GC4/ GC8/
CNTL SAC RECV GC-PTH4 PTH8
(n = 10) (n = 10) ( n = 10) ( n = 10) ( n = 10) ( n = 10) (n= 10)
Femur 15.5 15.4 15.3 15.6 15.5 15.6 15.4
Le ghmm) + 0.4 + 0.3 + 0.4 + 0.3 +0.2 +0.5 + 0.4
Results are reported as mean +standard deviation.










Bone Measures

Histomorphometric analysis was performed on the right femurs of 67 animals and the third

lumbar vertebra (L3) of 70 animals. One femur was damaged during tissue harvesting and could

not be sectioned and the other two were sectioned but could not be analyzed. Static and dynamic

parameters of bone change were measured or derived using methods described in Chapter 3.

Measured parameters using histomorphometry included BV/TV, Ob.S Oc.S, MS and MAR.

Once BV/TV was determined, values for Tb.N, Tb.Wi ', and Tb.Sp were derived using

calculations described elsewhere (51i). Once MS and MAR were measured, BFR/B S was

calculated as the product of these two. MicroCt was performed on 70 intact femurs and 70 intact

second lumbar vertebrae (L2). Static parameters of bone change measured using microCT

include BV/TV, Tb.N, Tb.Th", and Tb.Sp.

Data are presented as the measured or derived values for each parameter. There are also

tables showing percent change between groups that reached statistical significance. These tables

show the percent change compared to the group in the first column of the table. Percent change

is reported for all significant group interactions but only significant differences (p < 0.05) or

trends among comparable groups will be discussed. To simplify data presentation, a separate

table is provided within the chapter for each measured or derived parameter. Appendix A,

however, contains comprehensive tables for static, dynamic and cellular data by bone for each

measurement technique (histomorphometry and microCT).

Measurements of the Lumbar Vertebrae

Histomorphometry data for L3 are summarized in Table 4-4. Lumbar vertebrae L2 was

used for microCT and the results are shown in Table 4-5.


Histomorphometry analyzes a two-dimensional sample so the distance across trabeculae will be reported as Tb.Wi;
microCT analyzes samples in three dimensions so the distance across the trabeculae is reported as Tb.Th.










Table 4-4. Summary of Histomorphometric Analysis of LV3 by Group.
BSL CNTL 8 WKCNTL GC4/SAC GC 4/RECV GC 8 GC4/PTH4 GC-PTH8
(n = 10) (n = 10) (n = 10) (n = 10) (n = 10) (n = 10) (n = 10)
BV/TV 12.7 + 3.3 8.8+ 3.2 10.9 + 3.0 9.2+ 3.4 8.7 + 2.4 11.0 + 3.1 12.5 + 2.1
(%)

Tb.N 4.9 + 0.8 4.1+ 1.3 4.7 + 0.8 4.3 + 0.9 4.5 + 1.1 4.1 + 0.7 4.9 + 0.6
(1/mm)

Tb.Wi 31.0 + 4.6 25.3+ 5.0 27.5 + 3.8 25.1+ 5.1 23.2 + 3.0 31.5 + 4.3 30.9 + 4.9


Tb.Sp 185.6 + 38.4 242.1+ 82.4 196.7 + 42.4 223.0 + 6.4 214.5 + 6.1 220.7 + 39.6 182.3 + 21.0

Results are reported as mean +standard deviation.

Table 4-5. Summary of MicroCT Analysis of LV2 by Group.
BSLCNTL 8 WK CNTL GC4/SAC GC4/RECV GC 8 GC4/PTH4 GC-PTH8
(n= 10) (n = 10) (n = 10) (n = 10) (n = 10) (n = 10) (n = 10)
BV/TV (%/) 19.0 + 3.6 16.1 +3.6 17.4 + 4.3 15.0 + 3.8 19.0 + 4.6 18.8 + 4.8 22.0 + 4.4

Tb.N (1/mm) 4.2 + 0.3 3.9 + 0.5 3.9 + 0.5 3.8 + 0.5 4.4 + 0.8 3.8 + 0.5 4.04 + 0.4

Tb.Th (pm) 48.7 + 2.8 47.5 + 3.8 47.1 + 1.9 45.9 + 3.0 46.4 + 2.2 52.1 + 4.2 54.7 + 4.9

Tb.Sp (pm) 236 + 19 259 + 40 261 + 33 266 + 35 232 + 50 265 + 35 245 + 29

Results are reported as mean + standard deviation.

Bone Volume/Total Volume

Bone volume data based on histomorphometry are presented in Table 4.4, Table 4-6 and

Figure 4-1. Bone volume data based on microCT are presented in Table 4-5, Table 4-7, and

Figure 4-2. Bone volume was lower in older animals and higher in animals treated with PTH as

shown in Figure 4-1 and Figure 4-2. Histomorphometry indicated BV/TV was significantly

lower in the 8 WK CNTL (-30.7%, p = 0.03) group compared to BSL CNTL. Significant

increases in BV/TV were seen in animals treated with PTH. Animals treated with PTH for 8

weeks had a higher BV/TV than both the 8 WK CNTL (+42%, p = 0.01) and GC8 (+ 43.7%, p =

0.01) groups and a 35.9% (p = 0.03) higher BV/TV than animals in GC4/RECV. Animals in the

GC4/GC-PTH4 group had a 26.4% (p = 0.03) higher BV/TV than GC8 animals. MicroCT

analysis found significantly higher BV/TV in GC-PTH8 compared to 8 Wk CNTL (+36.6%, p =










0.01), GC4/SAC (+26.4%, p = 0.04), and GC4/RECV (+46.7%, p = 0.00). There was no bone

loss in GC8 compared to 8WK CNTL using either histomorphometry (8.7 +2.4 vs. 8.8 + 3.2) or

microCT (19.0 + 4.6 vs. 16.1 + 3.6). BV/TV did not increase after GC treatment ended.


Table 4-6. Significant Changes in Lumbar
Hi stomorphometry.
GROUP 8WK GC4/RECV
CNTL


Vertebra L3 Bone Volume by Group using


GC8


- 31.5%
(p=0.02)


GC4/GC-
PTH4


GC-PTH8


BSL CNTL

8 WK CNTL


GC4/RECV


- 30.7%
(p = 0.03)


- 27.7%
(p = 0.03)


+ 42.0%
(p = 0.01)

+ 35.9%
(p = 0.03)

+ 43.7%
(p = .01)


GC 8

Results are percent change compared to groups on the left.


+ 26.4%
(p = 0.03)


Bone Volume/Total Volume (BV/TV)


16
14
1-

12



0


BSL CNTL 8WK
CNTL


GC4/SAC GC4/RECV GC8


GC4/GC- GC-PTH8
PTH4


Figure 4- 1. Lumbar Vertebra L3 Bone Volume/Total Volume by Group using
Histomorphometry. a = significant compared to BSL CNTL; b = significant
compared to 8 WK CNTL; c = significant compared to GC4/RECV; d = significant
compared to GC8. Results are percent change compared to groups on the left.
Results are reported as mean + standard error.


a a a d b, ~d










Table 4-7. Significant Changes in Lumbar Vertebra L2 Bone Volume by Group using MicroCT.
Group GC4/RECV GC-PTH8


- 21.1%
(p = 0.03)


+ 36.6%
(p = 0.0 1)

+ 26.4%
(p = 0.04)

+ 46.7%
(p = 0.00)


BSL CNTL


8 WK CNTL


GC4/SAC


GC4/RECV


GC 8 21.1%
(p = 0.04)
Results are shown as percent change compared to groups on the left.


Bone Volume/Total Volume (BV/TV)


b,c,d


BSL 8WK GC4/ GC4/ GC8 GC4/GC- GC-PTH8
CNTL CNTL SAC RECV PTH4


Figure 4-2. Lumbar Vertebra L2 Bone Volume/Total Volume by Group using MicroCT. a =
significant compared to BSL CNTL; b = significant compared to 8 WK CNTL; c =
significant compared to GC4/SAC; d = significant compared to GC4/RECV; e =
si nifieant co pared to GC8. Results are re orted as mean + standard error.

Trabecular Number

Data for trabecular number are presented in Table 4-4, and Figure 4-3 (histomorphometry)

and Table 4-5 and Figure 4-4 (microCT). According to histomorphometry, the only significant

difference in trabecular number in L3 among treatment groups was between the GC4/GC-PTH4










and GC-PTH8, where the latter group had a higher (19.5%, p = 0.02) Tb.N. MicroCT, on the

other hand, showed only a non-significant 6.3% difference in these groups (Appendix A). There

was, however, a significant difference based on microCT between GC4/RECV (-9.5%, p = 0.04)

and GC4/GC-PTH4 (-9.5%, p = 0.02) compared to BSL CNTL.


Trabecular Number (Tb.N)


BSL 8WK GC4/ GC4/ GC8 GC4/GC- GC-PTH8
CNTL CNTL SAC RECV PTH4


Figure 4-3. Lumbar Vertebra L3 Trabecular Number by Group using Histomorphometry. a
si nificant co pared to GC4/GC-PTH4. Results are re orted as mean + standard
error.


Trabecular Number (Tb.N)


BSL 8WK GC4/ GC4/ GC8 GC4/GC- GC-PTH8
CNTL CNTL SAC RECV PTH4


Figure 4-4. Lumbar Vertebra L2 Trabecular Number by Group using MicroCT. a = significant
compared to BSL CNTL. Results are reported as mean + standard error.









Trabecular Width/Trabecular Thickness

Trabecular Width/Trabecular Thickness was a measure of the mean distance across the

trabeculae and changes reflected the effects of prednisolone, teriparatide, or both on this

distance. Data for Tb.Wi based on histomorphometry are found in Table 4-4, Table 4-8 and

Figure 4-5, while microCT data concerning Tb.Th are presented in Table 4-5, Table 4-9, and

Figure 4-6. As shown in Figure 4-5, histomorphometry generally detected higher Tb.Wi in PTH-

treated animals and lower Tb.Wi in older animals but no significant difference based on GC

treatment in comparably aged animals.

There was a significantly lower Tb.Wi between the BSL CNTL group and animals in the 8

WK CNTL (-18.4%, p = 0.02). Histomorphometry and microCT each found significant

differences between both of the PTH groups and 8 WK CNTL, GC4/RECV and GC8. There was

also a higher Tb.Wi in GC4/GC-PTH4 (24.5%, p = 0.01) and GC-PTH8 ( 22.1%, p = 0.01)

groups compared with comparably aged animals (8 WK CNTL). Additionally, animals treated

with PTH for either four or eight weeks had a 35.8% (p = 0.00) and 33.2% (p = 0.00) higher

Tb.Wi respectively compared to GC8 animals. There was no significant difference in Tb.Th

between GC4/GC-PTH4 and GC-PTH8.

Similar to histomorphometric Eindings, microCT showed significantly higher Tb.Th in

animals treated with PTH. Animals in the GC4/GC-PTH4 group had a greater Tb.Th than

animals in the 8 WK CNTL (+9.7%, p = 0.04), GC4/SAC (+10.6%, p = 0.01), GC4/RECV

(+13.5%, p = 0.00) and GC8 (+12.3%, (p=0.00) groups. Animals treated with PTH for eight

weeks showed a higher Tb.Th than 8 WK CNTL (+15.2%, p = 0.00), GC4/SAC (+16.1%, p =

0.00), GC4/RECV (+19.2%, p = 0.00), and GC8 (+17.9%, p = 0.00). Again, there was no

significant difference in Tb.Th between the GC4/GC-PTH4 and GC-PTH8 groups.




































b, c, b, d, e
40 d
aa
30-
25-


U| 1


BSL CNTL 8WK GC4/SAC GC4/RECV GC8 GC/GC- GC-PTH8


Table 4-8. Significant Changes in Lumbar Vertebra L3 Trabecular Width by Group using
Hi stomorphometry.
Group 8WK CNTL GC4/SAC GC4/RECV GC8 GC4/GC- GC-PTH8
PTH4


BSL
CNTL


8 WK
CNTL


- 18.4%
(p = 0.02)


- 19.0
(p=0.02)


- 25.1
(p= 0.00)


+ 24.5 %
(p =0.01)


+ 14.5%
(p = 0.03)

+ 25.5%
(p = 0.00)

+ 35.8%
(p = 0.00)


+ 22. 1
(p = 0.01)


GC4/SAC


GC4/RECV


+ 23.1%
(p = 0.01)

+ 33.2%
(p = 0.00)


GC 8


+ 18.5%
(p = 0.02)


Results are shown as percent change compared to groups on the left.


Trabecular Width (TIb Wi)


CNTL


PTH4


Figure 4-5. Lumbar Vertebra L3 Trabecular Width by Group using Histomorphometry. a =
significant compared to BSL CNTL; b = significant compared to 8 WK CNTL; c
significant compared to GC4/SAC; d = significant compared to GC4/RECV; e =
si nifieant co pared to GC8. Results are re orted as mean + standard error.





































tUU
70- b,c, a,b,c,
d,e d,e
S60
50


.l IIII


Table 4-9. Significant Changes in Lumbar Vertebra L2 Trabecular Thickness by Group using
MicroCT.


GC4/RECV

- 5.7%
(p = 0.02)


GC4/GC-PTH4


GC-PTH8

+ 12.3%
(p = 0.01)

+ 15.2%
(p = 0.00)

+ 16.1%
(p = 0.00)

+ 19.2%
(p = 0.00)

+ 17.9%
(p = 0.00)


+ 9.7%
(p = 0.04)

+ 10.6%
(p = 0.01)

+ 13.5%
(p = 0.00)

+ 12.3%


Group

BSL CNTL


8 WK CNTL


GC4/SAC


GC4/RECV


GC 8


(p = 0.00)


Results are shown as percent change compared to groups on the left.


Trabecular Thickness (Tb.Th)


BSL
CNTL


8WK GC4/ GC4/
CNTL SAC RECV


GC8 GC/GC- GC-PTH8
PTH4


Figure 4-6. Lumbar Vertebra L2 Trabecular Thickness by Group using MicroCT. a = significant
compared to BSL CNTL; b = significant compared to 8 WK CNTL; c = significant
compared to GC4/SAC; d = significant compared to GC4/RECV; e = significant
co pared to GC8. Results are re orted as mean + standard error.









Trabecular Separation

Trabecular separation data are shown in Table 4-4 and Figure 4-7 (histomorphometry) and

Table 4-5 and Figure 4-8 (microCT). Trabecular Separation is a reflection of the mean distance

between trabeculae within the region of interest. This distance tends to increase when resorption

is higher and decrease with increased bone formation. MicroCT results typically showed greater

trabecular separation for each group than did histomorphometric analysis, although the

differences were not significant. There were no significant differences in Tb.Sp related to age or

GC exposure based on microCT or histomorphometry. The only significant differences in

trabecular separation in L3 (histomorphometry) were found in PTH-treated animals. The GC-

PTH8 group had a significantly lower (-24.7%, p = 0.03) Tb.Sp compared with the 8 WK CNTL

and GC4/GC-PTH4 (-17.4%, p = 0.01) groups (Appendix A). MicroCT indicated significant

differences between BSL CNTL and GC4/RECV (+12.7%, p = 0.03) and GC-PTH8 (+12.3%, p

= 0.04).


Trabecular Separation (Tb.Sp)



250
L a,b
S200




50
BSL 8WK GC4/ GC4/ GC8 GC4/GC- GC-PTH8
CNTL CNTL SAC RECV PTH4

Figure 4-7. Lumbar Vertebra L3 Trabecular Separation by Group using Histomorphometry. a =
significant compared to 8 WK CNTL; b = significant compared to GC4/GC-PTH4.
Results are reported as mean + standard error.











Trabecular Separation (Tb.Sp)


350
;3 a a
S300






50
BSL 8WK GC4/ GC4/ GC8 GC4/GC- GC-PTH8
CNTL CNTL SAC RECV PTH4



Figure 4-8. Lumbar Vertebra L2 Trabecular Separation by Group using MicroCT. a =
si nificant co pared to BSL CNTL. Results are r ported as mean + standard error.

MicroCT methods showed significant changes in Tb. Sp only in relation to the BSL CNTL

group. Histomorphometry did not find changes in any group compared to the BSL CNTL group

but did find significant differences in animals treated with PTH, but only after 8 weeks of

treatment.

Measurements of the Distal Femur

The right femurs of 67 animals used for this study were harvested for histomorphometric

analysis while 70 left femurs from the same animals were analyzed using microCT. Three bones

could not be used for histomorphometric analysis for reasons previously described. The femurs

were harvested from each animal as described in Chapter 2. From each right femur, two 4 pum

thick sections were stained and used to measure or derive static bone measurements including

BV/TV, Tb.N, Tb.Wi, and Tb.Sp. The results of that analysis are summarized in Table 4-10.

Femurs used for microCT were scanned intact and the resulting measures of BV/TV, Tb.N,

Tb.Th, and Tb.Sp are presented in Table 4-11.











Table 4-10. Summary of Histomorphometric Analysis of the Distal Femur by Group.
BSL CNTL 8 WK CNTL GC4/SAC GC4/RECV GC 8 GC4/GC-PTH4 GC-PTH8
(n = 10) (n = 9) (n = 10) (n -= 10) (n= 9) (n= 9) (n= 10)
BV/TV (%) 4.8 +2.3 2.6 + 1.9 4.8 + 2.3 2.3 + 2.1 3.1 +1.6 4.5 + 1.7 5.2 + 2.9

Tb.N(1/nun) 2.0 + 0.9 1.2 + 0.8 1.8 + 0.9 1.2 + 0.9 1.6 + 0.8 1.9 + 0.7 2.2 + 0.9

Tb.Wi(pum) 28.2 + 5.9 25.4 + 7.2 26.1 + 5.9 21.9 + 5.5 23.2 + 6.1 27.9 + 4.3 28.6 + 6.1

Tb.Sp (pum) 540 + 225 1366+1146 694 + 428 1497 + 1812 904 + 916 551.6 + 195 554 + 359

Oc.S (%) 0.6 + 0.2 1.1 + 1.1 0.4 + 0.3 1.3 + 0.8 0.9 + 0.7 0.8 + 0.6 1.7 + 1.2

Ob.S (%) 7.9 +13.2 7.2 + 3.5 1.1 + 1.2 8.7 + 8.9 2.1 +1.5 13.6 + 12.0 20.0 + 7.8

MS (%) 6.7 +5.4 3.1 + 3.2 1.1 + 1.6 5.9 + 3.5 1.6 + 1.8 14.2 + 5.3 16.8 + 8.7

IVAR (pm/d) 0.9 +0.2 0.6 + 0.2 0.6 + 0.2 0.7 + .1 0.6 + 0.2 0.8 + 0.1 0.8 + 0.2

BFR/BS 6.4 + 6.0 2.2 + 3.0 0.6 + 1.0 4.2 + 2.7 1.0 + 1.1 10.8 + 5.7 14.8 +12.8
(um3/um2/d)
Results are reported as mean +standard deviation.

Table 4-11i. Summary of MicroCT Analysis of the Distal Femur by Group.
BSL CNTL 8WK GC4/SAC GC4/RECV GC 8 GC4/PTH4 GC-PTH8
(n = 10) CNTL (n = 10) (n = 10) (n = 10) (n = 10) (n = 10)
(n = 10)
BV/TV (%/) 9.9 + 3.1 6.2 + 2.1 8.5 + 3.4 6.0 + 3.8 8.7 + 2.6 7.9 + 2.9 11.5 + 3.1

Tb.N(1/mm) 2.7 + 0.63 2.1 + 0.45 2.5 + 0.57 2.2 + 0.78 2.8 +0.75 2.5 + 0.54 2.8 + 0.47

Tb.Th (pm) 60.7 + 5.2 64.5 + 6.3 60.1 + 2.8 59.9 + 6.4 56.3 +8.5 63.9 + 5.0 70.6 + 7.2

Tb.Sp (pm) 383 + 91 485 + 113 413 + 99 500 + 158 391 +154 416 + 108 361 + 60
Results are mean + standard deviation.

Bone Volume

Bone volume data based on histomorphometry are presented in Table 4-10, Table 4-12 and


Figure 4-9 while microCT data are presented in Table 4-11, Table 4-13, and Figure 4-10.

Histomorphometry and microCT detected the same trends in BV/TV in the distal femur.

Bone volume tended to be lower with age and higher with exposure to PTH. There was a

lower BV/TV in 8Wk CNTL compared to BSL CNTL in histomorphometry and microCT. GC-

PTH treatment resulted in higher BV/TV compared to 8 WK CNTL control and GC-treated

groups. Compared to GC/RECV, both groups receiving PTH had significantly increased bone

volume .










Table 4-12. Significant Changes in Distal Femur Bone Volume by Group using
Hi stomorphometry.
Group 8WK CNTL GC4/RECV GC4/GC-PTH4 GC-PTH8

BSL CNTL 45.8% 52.1%
(p = 0.03) (p = 0.01)

8WK CNTL +100.0%
(p = 0.02)

GC4/RECV + 95.6% + 126.1%
(p = 0.01) (p = 0.01)

Results are shown as percent change compared to groups on the left.


Bone Volume/T~Iotal Volume (BV/TV)

bc
6- a a c








BSL CNTL 8WK GC4/SAC GC4/RECV GC8 GC4/GC- GC-PTH8
CNTL PTH4


Figure 4-9. Distal Femur Bone Volume/Total Volume by Group using Histomorphometry. a =
significant compared to BSL CNTL; b = significant compared to 8 WK CNTL; c =
significant compared to GC4/RECV. Results are reported as mean + standard error.

There was only a slightly different BV/TV between GC4/SAC animals and the GC4/GC-

PTH4 and GC-PTH8 groups and these differences did not reach statistical significance. The GC-

PTH8 group had a higher BV/TV than GC4/GC-PTH4 (45.6%, p = 0.02) based on microCT.

MicroCT and histomorphometric comparisons of PTH groups with the 8 WK CNTL showed

BV/TV was higher in GC4/GC-PTH4 and the GC-PTH8 group, despite the fact these groups also

received GC.










Table 4-13. Significant Changes in Distal Femur Bone Volume by Group using MicroCT.


Group

BSL CNTL


8WK CNTL

- 37.4%
(p = 0.00)


GC4/RECV

- 39.4%
(p = 0.00)


GC8


GC-PTH8


8 WK CNTL


+ 40.3%
( p = 0.03)


+ 45.0%
( p =0.02)


+ 85.5%
(p = 0.00)


+ 91.7%
(p = 0.00)


+ 32.3%
(p = 0. 0.04


+ 45.6%
( p = 0.02)


GC4/RECV


GC 8


GC4/GC-PTH4


Results are shown as percent change compared to groups on the left.


BoneVolume/Total Volume (BV/TV)


16
1 4
12
10
o~8
6-
4
2
6


b, c,d, e


BSL
CNTL


8WK
CNTL


GC4/
SAC


GC4/
RECV


GC8 GC4/GC- GC-PTH8
PTH4


Figure 4-10. Distal Femur Bone Volume/Total Volume by Group using MicroCT. a =
significant compared to BSL CNTL; b = significant compared to 8 WK CNTL; c =
significant compared to GC4/RECV; d = significant compared to GC8; e = significant
compared to GC4/GC-PTH4. Results are reported as mean + standard error.















































I _^


Trabecular Number

Data for trabecular number based on histomorphometry are found in Table 4-10, Table 4-

14, and Figure 4-11. MicroCT data on this parameter are found in Table 4-11, Table 4-15, and

Figure 4-12. Histomorphometry and microCT detected similar patterns of change; where Tb.N

was lower in older animals and higher in animals treated with GC or GC-PTH.

Table 4-14. Significant Changes in Distal Femur Trabecular Number by Group using
Hi stomorphometry.


Group

BSL CNTL


8WK CNTL

- 40.0%
(p = 0.05)


GC4/RECV

- 40.0%
(p = 0.01)


GC-PTH8


8 WK CNTL


GC4/RECV


+ 83.3%
(p = 0.03)

+ 83.3%
(p = 0.03)


Results are shown as percent change compared to groups on the left.



Trabecular Number (Tb.N)


3
2.5

5 2


1
0.5


BSL
CNTL


8WK GC4/ GC4/
CNTL SAC RECV


GC8 GC4/ GC-
GC-PTH4 PTH8


Figure 4-11. Distal Femur Trabecular Number by Group using Histomorphometry. a =
significant compared to BSL CNTL; b = significant compared to 8 WK CNTL; c =
significant compared to GC4/RECV. Results are reported as mean + standard error.










Table 4-15. Significant Changes in Distal Femur Trabecular Number by Group using MicroCT.
Group 8WK CNTL GC8 GC-PTH8

BSL CNTL 22.2%
( p=0.05)

8 WK CNTL + 33.3% + 33.3%
(p = 0.03) (p = 0.00)


GC4/RECV + 27.3%
(p = 0.0 1)


Results are shown as percent change compared to groups on the left.


Trabecular Number (Tb.N)

3.5 bb,c







0.
BSL 8WK GC4/ GC4/ GC8 GC4/GC- GC-PTH8
CNTL CNTL SAC RECV PTH4

Figure 4-12. Distal Femur Trabecular Number by Group using MicroCT. a = significant
compared to BSL CNTL; b = significant compared to 8 WK CNTL; c = significant
co pared to GC4/RECV. Results are re orted as mean + standard error.

Trabecular Thickness/Trabecular Width

Data for Tb.Wi in the distal femur based on histomorphometry are found in Table 4-10,

Table 4-16, and Figure 4-13. MicroCT data for distal femur Tb.Th are found in Table 4-11,

Table 4-17, and Figure 4-14. Histomorphometry and microCT detected the same general

patterns in Tb.Th/Tb.Wi in the distal femur. Parameters of Tb.Wi and Tb.Th tended to be lower

in response to GC treatment and higher with PTH treatment. There was no significant age effect










apparent in Tb.Wi/Tb.Th as shown in Figure 4-13 and Figure 4-14. Both histomorphometry and

microCT found significant differences between PTH-treated animals and the 8 WK CNTL and

GC4/RECV groups. Additionally, there was a trend toward decreased Tb.N in GC4/SAC

compared to GC4/RECV in both histomorphometry.

Table 4-16. Significant Changes in Distal Femur Trabecular Width by Group using
Hi stomorphometry.
Group GC4/RECV GC8 GC4/GC-PTH4 GC-PTH8

BSL CNTL 22.3% 17.7%
(p = 0.04) (p = 0.05)


GC4/RECV


+ 27.4
(p = 0.05)

+ 20.3%
(p =0.04)


+ 30.6%
(p = 0.02)

+ 23.3%
(p = 0.03)


GC 8


Results are shown as percent change compared to groups on the left.



Trabecular Width (Tb.Wi)


35
30
S25
E 20
8 15
E 10
5-
0


b, c b, c


a ,


BSL
CNTL


8WK GC4/
CNTL SAC


GC4/
RECV


GC8 GC4/ GC-PTH8
GC-PTH4


Figure 4-13. Distal Femur Trabecular Thickness by Group using Histomorphometry. a =
significant compared to BSL CNTL; b = significant compared to GC4/RECV; c
si nifieant co pared to GC8. Results are re orted as mean + standard error.










Table 4-17. Significant Changes in Distal Femur Trabecular Thickness by Group using
MicroCT.
Group GC8 GC4/GC-PTH4 GC-PTH8


BSL CNTL

8 WK CNTL


+ 16.3%
(p = 0.01)


- 12.7%
(p = 0.04)


GC4/SAC


GC4/RECV


+ 6.3%
( p = 0.05)


+ 17.5%
(p = 0.00)

+ 17.9%
(p = 0.00)


+ 25.4%
(p = 0.00)


+ 10.5%
(p = 0.05)


GC 8


+ 13.5%
(p = 0.05)


GC4/GC-PTH4


Results are shown as percent change compared to groups on the left.


Trabecular Thicknless (Tb.Th)


b cle d,e,f


BSL
CNTL


8WK GC4/ GC4/ GC8 GC/GC- GC-PTH8
CNTL SAC RECV PTH4


Figure 4-14. Distal Femur Trabecular Thickness by Group using MicroCT. a = significant
compared to BSL CNTL; b = significant compared to 8WK CNTL; c = significant
compared to GC4/SAC; d = significant compared to GC4/RECV; e = significant
compared to GC8; f= significant compared to GC4/GC-PTH4. Results are reported
as mean + standard error.











































Results are shown as percent change compared to groups on the left.


Trabecular Separation (Tb.Sp)


Trabecular Separation

Data for trabecular separation based on histomorphometry are found in Table 4-10, Table

4-18, and Figure 4-15. Data from microCT for Tb.Sp are found in Table 4-11, Table 4-19, and

Figure 4-16. In general, higher Tb.Sp was seen in older animals and both GC and GC-PTH

tended to decrease this parameter as shown in Figure 4-15 and Figure 4-16.

Compared to the BSL CNTL group, Tb.Sp was higher (+153.0%, p = 0.05) in the 8WK CNTL

group based on histomorphometry.

Table 4-18. Significant Changes in Distal Femur Trabecular Separation by Group using
Hi stomorphometry.


Group

BSL CNTL


8 WK CNTL


GC4/RECV


8WK CNTL

+153.0%
(p = 0.05)


GC4/RECV

+ 177.2%
(p = 0.01)


GC4/GC-PTH4


GC-PTH8


- 59.4%
(p = 0.03)

- 63.0%
(p = 0.03)


- 63.3%
(p = 0.01)


S20(

E 15(
o
.Y 10(

5(


c b, c


BSL
CNTL


8WK GC4/
CNTL SAC


GC4/
RECV


GC8 GC4/GC- GC-PTH8
PTH4


Figure 4-15. Distal Femur Trabecular Separation by Group using Histomorphometry. a =
significant compared to BSL CNTL; b = significant compared to 8 WK CNTL; c =
significant compared to GC4/RECV. Results are reported as mean + standard error.










Table 4-19. Significant Changes in Distal Femur Trabecular Separation by Group Using
MicroCT.
Group GC4/RECV GC8 GC-PTH8

BSL CNTL + 30.5%
(p = 0.05)
8 WK CNTL 19.4% 25.6%
(p = 0.03) (p = 0.00)

GC4/RECV 27.8%
(p = 0.01)

Results are shown as percent change compared to groups on the left.


Trabecular Separation (Tb.Sp)



600 b b,c


E 500




BSL 8WK GC4/ GC4/ GC8 GC4/GC- GC-PTH8
CNTL CNTL SAC RECV PTH4

Figure 4-16. Distal Femur Trabecular Separation by Group using MicroCT. a = significant
compared to BSL CNTL; b = significant compared to 8 WK CNTL; c = significant
compared to GC4/RECV. Results are reported as mean + standard error.

Osteoblast Surface and Osteoclast Surface

Osteoblasts and osteoclasts are the cells involved in resorbing and forming new bone.

They act in concert and the balance of numbers and activity levels of these two cell types

determines whether there is an overall increase or decrease in bone. Many disease conditions

affect the numbers and activity levels of these cells. The ratio in number rand activity levels are

also influenced by treatments, including PTH. Osteoblast and Osteoclast Surface data are

presented in Table 4-10, Table 4-20, and Figures 4-17 and 4-18 There were no significant age-










related changes in Ob.S or Oc.S in the distal femur as shown in Figure 4-17 and Figure 4-18.

Ob.S tended to decrease with GC and increase with GC recovery or PTH. Trends in Oc.S were

not as clear cut.

Table 4-20. Significant Changes in Distal Femur Osteoblast Surface and Osteoclast Surface by
Group .


Group
BSL CNTL


GC4/SAC GC4/RECV
Oc.S
+ 116.7%
(p = 0.02)


GC8


GC4/GC-PTH4


GC-PTH8
Oc.S
+ 183.3%
(p = 0.0 1)

Ob.S
+ 153.2%
(p = 0.0 1)

Ob.S
+ 177.8%
(p =0.00)


8 WK CNTL


Oc.S
- 63.6%
(p = 0.03)

Ob.S
- 84.7%
(P= 0.00)


Ob.S
-70.8
(p = 0.00)


GC4/SAC


Oc.S
+ 225%
(p = 0.00)

Ob.S
+ 690.9%
(p = 0.00)


Ob.S
+1,136.4%
(p= 0.00)


Oc.S
+ 325.0%
(p = 0. 00)

Ob.S
+ 1,718.2%
(p = 0.00)

Ob.S
+ 129.9%
(p = 0.00)

Ob.S
+ 852.4%
( p= 0.00)


Oc.S
+ 112.5%
(p = 0.03)
Ob.S
+ 47.1%
(p = 0.05)


GC4/RECV


GC 8


Ob.S
+ 314.30 %
(p = 0. 00)


Ob.S
+ 547.6%
(p = 0.00)


GC4/GC-PTH4


Results are shown as percent change compared to groups on the left.














Osteoblast Surface (Ob.S)


30
25

S20
15
10
5
0


a, b, c
d, e, f


b, d


BSL
CNTL


8WK
CNTL


GC8 GC4/GC- GC-PTH8
PTH4


GC4/ GC4/
SAC RECV


Figure 4-17. Distal Femur Osteoblast Surface by Group using Histomorphometry. a =
significant compared to BSL CNTL; b = significant compared to 8 WK CNTL; c =
significant compared to GC4/SAC; d = significant compared to GC4/RECV; e =
significant to GC8; f = significant compared to GC4/GC-PTH4. Results are reported
as mean + standard error.




Osteoclast Surface (Oc.S)


a,c, d
T


BSL
CNTL


8WK
CNTL


GC8 GC4/GC- GC-PTH8
PTH4


Figure 4-18. Distal Femur Osteoclast Surface by Group using Histomorphometry. a =
significant compared to BSL CNTL; b = significant compared to 8 WK CNTL; c = significant
compared to GC4/SAC; d = significant compared to GC4/GC-PTH4. Results are reported as
mean + standard error.


GC4/ GC4/
SAC RECV










Dynamic Measures of Bone Formation in the Distal Femur

Dynamic measures included mineralizing surface (MS), mineral apposition rate (MAR)

and surface referent bone formation rate (BFR/B S). Changes in bone formation were measured

or calculated using unstained 8 Clm-thick sections. Data for MS, MAR, and BFR/BS are

provided in Table 4-10. Differences in MS, MAR, and BFR/BS between groups are shown in

Tables 4-22, Table 4-23 and Table 4-24 as well as in Figure 4-19, Figure 4-20 and Figure 4-21,

respectively.

Mineralizing Surface

Mineralizing surface was determined by measuring the total perimeter of trabeculae and

the percentage of the perimeter with double fluorochrome labeling. Significant changes in MS

are shown in Table 4-21.

Mineralizing surface was lower in older animals and those exposed to GC but tended to

increase when GC was discontinued and when PTH was used as shown in Figure 4-19. Animals

in GC8 had 48.4% (p = 0.05) less MS than 8 WK CNTL, while GC4/SAC had 64.5% (p = 0.02)

less MS than did 8 Wk CNTL. When GC treatment was discontinued, MS increased. Compared

to GC4/SAC, GC4/RECV had 436.4% (p = 0.00) higher MS and GC4/RECV had a 268.8%,p =

0.00) higher MS than did GC8. GC4/GC-PTH4 also showed a higher MS that 8 WK CNTL

(+358.1%, p = 0.00), GC4/SAC (1,190.0%, (p = 0.00), GC4/RECV (+140.7%, p = 0.01), and

GC8 (787.5%, p = 0.00). Mineralizing surface was even more pronounced in the GC-PTH8

group where it was higher than 8 WK CNTL (441.9%), p = 0.00), GC4/SAC (1,427.3%, p =

0.00), GC4/RECV (184.7%, p = 0.00), and GC8 (950.0%, p = 0.00).

MAR was lower in older animals but was not significantly decreased by GC use.

However, MAR tended to increase when GC was discontinued and when PTH was administered












Table 4-21. Significant Changes in Distal Femur Mineralizing Surface by Group.
Group GC4/SAC GC4/RECV GC8 GC4/GC-PTH4 GC-PTH8


BSL CNTL


8 WK CNTL


- 83.8%
(p = 0.01)

- 65.4%
(p = 0.02)


- 76.8%
(p = 0.02)

- 48.4%
(p = 0.05)


+111.9%
(p = 0.01)

+ 358.1%
(p= 0.00)

+ 1,190.0%
(p = 0.00)


+ 140.7%
(p = 0-.01)

+ 787.5%
(p = 0.00)


+ 150.7%
(p = 0.00)

+ 441.9%
(p = 0.00)

+1,427.3%
(p = 0.00)


+ 184.7%
(p = 0.00)

+950.0%
(p = 0.00)


+ 90.3%
(p = 0.05)

+ 436.4%
(p = 0.00)


GC4/SAC


GC4/RECV


GC 8


+268.8%
(P = 0.00)


Results are shown as percent change compared to groups on the left.


Mineralizing Surface (MS)


30
25
20

15
10


a~b,c,d


ab b, c, e


BSL CNTL 8WK GC4/SAC GC4/RECV GC8 GC4/GC- GC-PTH8
CNTL PTH4


Figure 4-19. Distal Femur Mineralizing Surface by Group using Histomorphometry. a =
significant compared to BSL CNTL; b = significant compared to 8 WK CNTL; c
significant compared to GC4/SAC; d = significant compared to GC4/RECV; e=
significant compared to GC8. Results are reported as mean + standard error.













































I


Mineral Apposition Rate


Significant changes in MAR are shown in Table 4-22 and Figure 4-20.


Table 4-22.


BSL CNTL


Significant Changes in Distal Femur Mineral
8WK CNTL GC4/SAC GC4/RECV


Apposition Rate by Group.
GC8 GC4/GC-
PTH4
33.3%
(p =
0.02)
+ 33.3%
(p = 0.01)


GC-PTH8


- 33.3%
(p = 0.01)


-33.3%
(p = 0.01)


+ 16.7%
(p = 0.03)

+ 16.7%
(p = 0.03)


+ 33.3%
(p = 0.02)

+ 33.3%
(p = 0.02)


8 WK CNTL


GC4/SAC


GC4/RECV


GC 8


+ 33.3%
(p = 0.01)

+ 14.3%
(p = 0.02)

+ 33.3%
(p = 0.02)


Results are shown as percent change compared to groups on the left.


Mineral Apposition Rate (MAR)


1.1
0.9
0.7
S0.5
0.3
0.1
-0.1


a b, c, d,e c


a,d a,d


BSL 8WK GC4/ GC4/ GC8 GC4/GC- GC-PTH8
CNTL CNTL SAC RECV PTH4


Figure 4-20. Distal Femur Mineral Apposition Rate by Group using Histomorphometry. a
significant compared to BSL CNTL; b = significant compared to 8 WK CNTL; c
significant compared to GC4/SAC; d = significant compared to GC4/RECV; e =
significant compared to GC8. Results are reported as mean + standard error.











as shown in Figure 4-20. Mineral Apposition Rate was 33.3% (p = 0.01) lower in 8 WK CNTL

than in BSL CNTL. Mineral Apposition rate remained the same between GC8 and 8 WK CNTL

(0.6 Clm/day). However, in animals recovering from GC exposure, GC4/RECV had a 16.7% (p =

0.03) higher MAR than 8 WK CNTL and a 16.7% (p = 0.03) higher MAR compared with the

GC4/SAC group.

Exposure to PTH resulted in the GC-PTH groups having significantly higher MAR than

control or GC-treated groups. Compared with comparably aged animals in the 8 WK CNTL

group, MAR in the GC4/GC-PTH4 group was 33.3% (p = 0.01) greater. The MAR in GC/GC-

PTH4 group was also 33.3% (p = 0.01) higher than in the GC4/SAC group, 14.3% (p = 0.02)

higher than in GC4/RECV, and 33.3% (p = 0.02) higher than in GC8. MAR was greater in GC-

PTH8 than in 8 WK CNTL (+33.3%, p = 0.02)

Bone Formation Rate/Bone Surface (BFR/BS)

Both MS and MAR are measured directly, while the third dynamic measure in bone,

BFR/B S, was calculated. BFR/BS is derived by multiplying MS x MAR and represents the

average amount of bone formed daily (um3/um2/day). Significant changes in BFR/B S based on

treatment groups are found in Table 4-23. There was a non-signifieant trend toward lower

BFR/BS in older animals and a significant decrease in BFR/BS following GC exposure in the

GC4/SAC group compared with the 8 Wk CNTL group. However, there was no significant

difference in BFR/BS between the GC8 and 8 WK CNTL group. PTH treatment did result in

significantly higher BFR/B S. The GC4/PTH4 group had a BFR/B S higher than the 8 WK CNTL

(+390.9%, p = 0.01), the GC4/SAC (+1,700%, p = 0.00), the GC4/RECV (+157. 1%, p = 0.02),

and the GC8 (+980%, p = 0.00) groups. This pattern continued with the GC-PTH8 group.










Table 4-23. Significant Changes in Distal Femur Bone Formation Rate by Group using
Hi stomorphometry.
GC4/SAC GC4/RECV GC8 GC4/GC-PTH4 GC-PTH8
BSL CNTL 90.6% 84.4% + 131.3%
(p = 0.00) (p = 0.01) (p = 0.02)


- 72.7%
(p = 0.02)


+ 90.9%
(p = 0.03)

+ 600.0%
(p = 0.00)


+ 390.9%
(p = 0.01)

+ 1,700%
(p = 0.00)

+ 157.1%
( p = 0.02)

+ 980.0%
(p = 0.00)


+ 572.7%
( p = 0.00)

+ 2.366.7%
(p = 0.00)

+ 254.4%
(p = 0.00)

+ 1,380.0%
(p = 0.00)


8 WK CNTL


GC4/SAC


GC4/RECV


GC 8


+ 76.2%
(p = 0.00)


Results are shown as percent change compared to groups on the left.



Bone Formation Rate/Bone Surface (BFR/BS)


a,b, c,
b,c,d,e d,e


c6 20
S15
10


a,b b,c,e


BSL 8WK GC4/ GC4/ GC8 GC4/
CNTL CNTL SAC RECV GC-
PTH4


GC-PTH8


Figure 4-21. Distal Femur Bone Formation Rate/Bone Surface by Group using
Histomorphometry. a = significant compared to BSL CNTL; b = significant
compared to 8 WK CNTL; c = significant compared to GC4/SAC; d = significant
compared to GC4/RECV; e = significant compared to GC8. Results are reported as
mean + standard error.










Mid-Shaft Cortical Bone Data and Significant Changes

A portion of the mid-shaft of the left femur in 70 mice was also analyzed using microCT. In

each animal, 55 slices (1.2 Clm each) were scanned and reconstructed. Cortical thickness (Ct.Th)

data are presented in Table 4-24, Table 4-25, and Figure 4-22. There was no apparent age-

related difference in cortical thickness. While Ct.Th did not change significantly with GC

treatment, it did increase significantly with PTH as shown in Figure 4-22. Animals receiving 8

weeks of GC and PTH had significantly higher Ct.Th than BSL CNTL (+8.7%, p = 0.02), GC8

(9.5%, p = 0.02), or GC4/SAC (+10.7%, p = 0.00). The only other significant difference

occurred where Ct.Th was 6.7% higher (p = 0.03) in the GC4/GC-PTH4 group than in GC4/SAC

animals.

Table 4-24. Summary of Histomorphometric Analysis of Femur Mid-Shaft by Group.
BSL 8 WK GC4/ GC4/ GC 8 GC4/GC- GC-
CNTL CNTL SAC RECV (n =10) PTH4 PTH8
(n = 10) (n= 10) (n = 10) (n = 10) (n = 10) (n = 10)
Cortical 27.5 + 1.9 28.3 +2.4 27.0 + 1.4 28.2 + 1.3 27.3 + 1.7 28.8 + 1.7 29.9+2.0
Thickness
Results are reported as mean +standard deviation.



Table 4-25. Summary of Signifieant Changes in Femur Mid-Shaft Cortical Bone Thickness by
Group using MicroCT.
GC4/GC-PTH4 GC-PTH8
BSL CNTL + 8.7%
(p = 0.02)

GC8 9.5%
(p = 0.02)

GC4/SAC4 + 6.7% 10.7%
(p = 0.03) (p = 0.00)
Note: Results are shown as percent change compared to groups on the left.











Cortical Thickness (Ct.Th)


BSL 8WK GC4/ GC4/ GC8 GC4/GC- GC-PTH8
CNTL CNTL SAC RECV PTH4


Figure 4-22. Mid-Shaft Femur Cortical Thickness by Group using MicroCT. a = significant
compared to BSL CNTL; b= significant with GC8; c = significant with GC4/SAC
Results are reported as mean + standard error.









CHAPTER 5
DISCUSSION

This was the first prospective study to evaluate the ability of PTH treatment to inhibit the

negative skeletal effects of GC in mice. The study examined several aspects of this issue

including the cumulative effects of both GC and PTH over time, site specific responses, tissue

and cellular changes in response to treatment, and the effectiveness of natural recovery from

termination of GCs. The maj or findings of this investigation are as follows:

* GCs suppress bone formation/turnover but do not necessarily reduce bone volume

* PTH was bone anabolic even in the presence of GCs

* PTH increased bone mass quickly

* There was a residual effect following discontinuation of GC therapy that resulted in no
increase in bone mass despite a rebound effect in osteoblast numbers and activity

* Following GC treatment with PTH was more effective in improving bone structural
parameters and mineralization than was natural recovery

* The magnitude of response to GC and PTH varied by skeletal site

* There was an age-related decline in bone structural parameters in control animals.

Glucocorticoid Drugs Suppressed Bone Formation But Did Not Affect Bone Volume

A previous study found that bone turnover decreased 67.4% (p = 0.05) in Swiss Webster

mice treated with GC for 4 weeks (5). Other studies have reported similar findings (2,5,114) and

it is generally accepted that exposure to GCs suppresses bone turnover in mice. We also found

that turnover decreased but, in contrast to some studies, we observed increased BV/TV, though

this only reached the level of statistical significance in the distal femur and only in measurements

using microCT.

GCs may have caused an uncoupling of the remodeling process so there was no longer the

same amount of bone formed as resorbed. In some previous studies this resulted in a lower









BV/TV in mice, but in the present study we detected a bone specific increase in BV/TV. While

we found no significant changes in lumbar spine BV/TV after 8 weeks of GC use, we did detect

(by microCT) a significant increase in BV/TV in the distal femur.

At the cellular level, GCs affect osteoblasts, osteocytes, and osteoclasts but typically affect

the osteoblast most (39). GCs typically decrease osteoblast and osteoclast numbers and activity

levels while also increasing osteocyte apoptosis. The effects on osteoclasts are more variable.

Studies invariably find Ob.S declines, but studies have found both increases and decreases in

Oc.S (39). Lane found significant increases in Oc.S while Weinstein found increased and

decreased Oc. S in different studies (2,5,114). In our study, 4 weeks of GC resulted in an 84.7%

(p = 0.03) lower Ob.S and a 63.6% (p = 0.03) lower Oc.S in the distal femur in GC-treated

animals.

In the present study, decreased Ob.S had far-reaching consequences. Mineralizing surface

decreased by over 60%, further suggesting an important effect of GCs on osteoblast numbers.

Mineralizing surface generally reflects the number of osteoblasts on trabecular surfaces and

histomorphometry detected a trend toward decreased Ob.S and Oc.S in the GC8 group compared

to 8 WK CNTL although the changes did not reach statistical significance. It is possible that the

increased BV/TV seen in this study results from GCs having more of a suppressive effect on

osteoclasts than on osteoblasts. If osteoclast activity levels were suppressed more, relative to

osteoblast levels, this would result in increased bone volume.

GCs not only affect bone cell numbers, they also influence the activity levels of these cells.

In osteoblasts, GCs reduce the time these cells secrete osteoid and actively mineralize bone (72).

Although we did not measure osteoid, measurement of dynamic parameters suggest a decrease in

bone turnover. In the distal femur, we found MS was 48.4% (p = 0.05) lower in GC-treated









animals (GC8) than in 8 WK CNTL animals but found no difference in MAR, an index of

osteoblast activity, and only a non-signifieant, two-fold decrease in BFR/BS. These Eindings are

similar to other studies with respect to MS but differ with studies that found lower MAR and

BFR/B S (2,5). Our findings with respect to MAR and BFR/BS are consistent with the

differences in BV/TV we detected, however.

It is more difficult to directly assess osteoclast activity levels, but inferences can be made.

The typical response to GCs in human disease is decreased bone turnover that disproportionately

affects formation rather than resorption. This results in rapid bone loss. Most studies of GC in

mice have found that BV/TV declines in a dose-dependent and time-dependent manner. This

was the case in several studies, including some using Swiss Webster mice A previous study

found that 6-month old Swiss Webster mice treated for three weeks had a modest 19% (p = 0.05)

lower BV/TV in the lumbar vertebrae based on histomorphometry and 22% (p = 0.05) decrease

based on microCT (2. In this same stud Oc.S increased over 100 %, from 0.8 to 1.7% (p

0.05 MS decreased by almost one-third, from 42.9 to 29.5% (p 0.05) while MAR declined

almost 40% from 1.03 to 0.64 Cpm (p < 0.05). These changes accompanied a 105% increase in

osteocyte apoptosis from 1.9 to 3.9% (p < 0.05) (2). Findings such as these have previously been

attributed to a decrease in osteoblastogenesis and increased apoptosis of both osteoblasts and

osteocytes (2,5,39).

What determines the net balance of formation to resorption is unclear, but our Einding of

increased BV/TV in mice should not be dismissed without further study since the same

phenomenon has been seen in other rodents. In rats, researchers have detected bone loss, bone

increase, or no change after exposure to GCs (36-3 8). There has also been variability reported in

mice exposed to GCs. While some studies in mice have reported dose- and time-dependent









losses in BV related to GC use, others have not. Using histomorphometry, one study found no

significant decline in BV/TV even though BFR/BS declined significantly (~75%) in the femur.

The only significant differences detected by microCT occurred at very high GC dosages, which

are known to impact other steroid receptors (3). This suggests that the dose-response to GCs is

not linear and that some conditions may cause resorption to be more suppressed than formation.

Our Eindings suggest, that while changes in bone cell numbers were important, changes in

activity levels at the cellular level may also be central to the changes we detected in these bones.

While both osteoblasts and osteoclasts decreased in number in response to GC treatment, the

increased BV/TV found with GC treatment found in our study suggests osteoclast activity levels

may have been more severely suppressed relative to osteoblast activity, allowing for increased

BV/TV despite decreased numbers of osteoblasts. This requires further examination, however,

since there were no measures in this study to detect or quantify osteoclast activity levels.

Anabolic Effects of PTH Prevented the Inhibitory Changes Associated with Glucocorticoid
Drugs

Studies evaluating the effects of PTH on mice consistently show it is bone anabolic even

under conditions that tend to decrease bone formation and/or bone mass (7,8,45) and our study

supports this finding as well. In one study involving ovariectomized mice, PTH treatment

resulted in a two-fold increase in MAR and a 3-fold increase in BFR/BS (7).

GCs suppress osteoblast proliferation/ activity levels and PTH, a hormone that stimulates

osteoblasts, is able to reverse this trend (6) to overcome the suppressive effects of GCs. PTH

and GC together have been tested in rats (115), but the present study is the first to examine the

effects of PTH in GC-treated mice. The only human investigation of the combined effects of GC

and PTH involved subjects who also took estrogen. In that study, BMD increased linearly

during the entire 12 month course of treatment (6). This finding was consistent with other









studies that found that the antiresorptive effects of estrogen did not prevent the bone anabolic

response to PTH (116).

The anabolic nature of intermittent PTH on bone is seen in its effects at the cellular level.

Following PTH administration to rats and humans, there is typically an increase in both Ob.S and

Oc.S, leading to increased bone turnover. Although both Oc.S and Ob.S increase, this change is

most dramatic in Ob.S and, therefore, favors bone formation. A study of PTH in intact mice

reported a significant increase in Ob.S (45). We found no significant change in Oc.S but a

significant increase in Ob.S compared to the GC8 group after 4 weeks (+ 547.6%, p = 0.00) and

8 weeks (+ 852.4%, p = 0.00) of PTH treatment. This substantial change in osteoblast presence

on trabecular surfaces could reflect increased differentiation, decreased apoptosis, or reactivation

of bone lining cells. Consistent with the increase in Ob.S, after 4 weeks of treatment with PTH,

there was a significant increase in MS (787.5%, p = 0.00) MAR (33.3%, p = 0.02), and BFR/BS

(980.0%, p = 0.00) compared to animals receiving eight weeks of GC. The differences in MS

and BFR/BS were even more pronounced after 8 weeks of PTH.

In this study, Ob.S was over 1,000% higher in a group of animals treated with PTH for 4

weeks compared to animals treated similarly with GCs (GC4/SAC) The increase in number

and activity levels of the osteoblasts resulted in a BFR/B S over ten times the rate seen in GC8

animals in just 4 weeks. Likewise, animals treated with PTH and GC for 8 weeks had over

850% greater Ob.S than did GC8 animals. The magnitude of change in Ob.S makes it unlikely

that it could result solely from reactivation of quiescent bone lining cells. The very high

increases in Ob.S seen with PTH likely stem from increased osteoblast proliferation as well as

increases in osteoblast lifespan (116).









Mineralizing surface is an indirect measure of the osteoblast population and increased

numbers of these cells means more bone surface is mineralizing at any given time.

Mineralization apposition rate generally reflects the activity level of the osteoblasts, since

osteoblasts modulate bone mineralization. The significantly higher MS and MAR detected in the

present study are evidence PTH increased both the number and activity levels of osteoblasts.

BFR/B S, a function of both MS and MAR, is also significantly higher with PTH, reflecting the

shifting balance in bone turnover toward bone formation. There are only a few studies that have

examined the effects of PTH following GC therapy on BFR/BS. One of these found that

BFR/B S tended to decrease in response to GC alone but increased with PTH alone in rats. That

study found that when the two drugs are used simultaneously, there was an intermediate response

that increases BFR/BS relative to GC-only treated animals, but kept it lower than with PTH

alone (115).

The effects of PTH also appeared to increase over time. Animals in the GC-PTH8 group

showed a trend toward even higher bone mass than the GC4/GC-PTH4 animals, according to

microCT, but high variability prevented this from reaching statistical significance. MicroCT

found a 45.6% (p = 0.02) greater BV/TV in distal femur between animals treated with PTH for 8

weeks versus those treated for only 4 weeks. These changes appear driven by additional

increases in osteoblast and osteoclast numbers since Ob.S was 47.1% (p = 0.05) higher and Oc.S

was 112.5% (p = 0.03) higher in group receiving 4 additional weeks of PTH. At the end of the 8

weeks of PTH treatment, osteoclasts lined 1.7% of the trabecular surfaces while osteoblasts lined

20% of the trabecular surfaces.

PTH Increases Bone Mass Quickly

Studies have found the effects of PTH are rapid and one study found significant changes in

BMD in the tibia of C57BL/J6 mice with just one week of treatment (45), although it took longer









to detect BMD differences in the vertebrae. In the present study, there was also significantly

higher Ob.S, MS, MAR, and BFR/BS in the distal femur after 4 weeks of PTH-treatment

indicating increases in osteoblast numbers and activity levels quickly influenced both bone

formation and mineralization activity. These improvements were evident in increased vertebral

BV/TV and Tb.Wi at 4 weeks based on histomorphometry and in Tb.Th based on microCT.

Our study mirrors other studies in finding early and significant changes in dynamic indices of

bone turnover but a lag of at least 4 or more weeks before structural differences reach statistical

significance difference from baseline controls (8,45).

Changes in bone structural parameters were brought about through the ability of PTH to

increase bone turnover even though GC was also administered. Animals receiving PTH for 4

weeks showed increases in Ob.S (+547.6%, p = 0.00) but a non-significant decrease in Oc.S

(-11.1%, p = ns) compared to GC8 animals. This helps show why PTH is anabolic since the

proportion of the trabecular surfaces covered in osteoblasts increased over 5-fold. PTH also

increased the activity levels of osteoblasts as reflected in higher MS, MAR, and BFR/BS. Of

interest, animals treated with PTH for 8 weeks while also receiving GCs, showed a further

increase in Ob.S (+47.1%, p = 0.05) and Oc.S (+112.5%, p = 0.03) compared to animals in the

GC4/GC-PTH4.

Residual Effects of Glucocorticoid Drugs are Apparent During Natural Recovery

No previous studies have evaluated bone recovery once GC treatment is discontinued. We

expected discontinuation of GC treatment would result in natural recovery that would bring the

measured parameters closer to control animals. That was not the case since we detected trends

toward lower BV/TV, Tb.N and Tb.Th and higher Tb.Sp in the GC4/RECV group than in the

GC4/SAC, a group of animals sacrificed immediately following four weeks of GC treatment.

This suggests there is a residual effect to GC use that, at least in mice, does not abate in four










weeks. Although there were no statistically significant changes in bone structural parameters,

there were significant changes in activity at the cellular level. Four weeks of recovery resulted in

significantly higher Oc.S (+225%, p = 0.00) and Ob.S (+ 690.9%, p = 0.00) compared to

GC4/SAC. There was a "rebound" effect in dynamic bone measures, that resulted in MS, MAR,

and BFR/BS exceeding even the 8 WK CNTL group. Increases in these parameters, however,

did not translate into increased bone mass, suggesting either there was not sufficient time for

increased bone mass to be detected or there were GC-induced changes to the bone matrix

preventing expected increases. We believe the latter is more likely and that this may represent a

change to the bone matrix that inhibits increases in structural bone mass even once GC treatment

is discontinued.

The reason for the impaired ability of these bones to increase in mass despite high activity

levels in bone cells is unclear, but is not without precedent. Experiments in animals treated with

alcohol show similar impaired recovery. When decalcified bone cores from alcohol-treated and

control mice were placed subcutaneously in untreated mice, the alcohol-exposed cores showed

an impaired ability to re-mineralize (117). It is possible a similar process is at work in

glucocorticoid-treated mice. Conversely, it is also possible a longer recovery period would have

made a difference. After GCs are discontinued in humans there is a slow reduction in fracture

risk over time (23). One study found the relative fracture risk decreased from 2.4 to 1.8 one year

after GC is stopped, and continued to move back to pre-treatment baseline risk levels over time

(23). These reductions in fracture risk may result from improved structural repairs to the bone

over time.

PTH Treatment After Glucocorticoid Use Was More Effective than Natural Recovery

We expected administration of PTH following GC treatment to be more effective in

increasing bone structural parameters than natural recovery, and the data support this. In the









vertebrae, Tb.Th was increased by 25.5% (p = 0.00) using histomorphometry and by 13.5% (p =

0.00) using microCT in GC4/GC-PTH4 compared with GC4/SAC. In the distal femur, BV/TV

and Tb.Wi were significantly higher in GC4/GC-PTH4 compared with GC4/SAC while Tb.Sp

was significantly lower based on histomorphometry. In the distal femur, BV/TV increased

significantly (95.6% (p = 0.01) although there was only a non-significant increase in Ob.S

between the GC4/GC-PTH4 and GC4/SAC groups. Histomorphometry did, however, detect

significant increases in MS (140.7%, p = 0.01), an indication that a greater percentage of the

trabeculae were covered in osteoblasts. There was also a 14.3% (p = 0.02) increase in MAR, a

measure of osteoblast activity levels. The increases in MS and MAR contributed to the

significant increase in BFR/BS (+ 157.1, p = 0.02) indicating that PTH treatment resulted in

increased bone formation. These findings were expected; since PTH in bone anabolic we

expected it would have a greater effect on bone parameters than natural recovery after GC-

treatment was discontinued. While animals that experienced natural recovery showed a rebound

effect in dynamic bone parameters that showed significant increases in MS(90.3%, p = 0.05),

MAR (16.7%, p = 0.03), and BFR/BS (90.9%, p = 0.03) compared to 8 WK CNTL, these levels

were much more impressive following PTH treatment: MS (441.9%, p = 0.00), MAR (33.3%, p

= 0.02), BFR/BS (572.7%, p = 0.00).

Age-Related Effects on Bone Mass

We expected a relatively constant level of bone turnover in these adult male Swiss Webster

mice. A previous study, whose goal was specifically to determine when long-bone growth

ceased, found that this strain of mice cease longitudinal bone growth in the long bones between

five and six months of age (5). Based on this, we expected young adult Swiss Webster mice to

have a constant level of bone turnover that would result in stable levels of bone mass throughout

the two-month length of the study. Our data suggest, however, this was not the case. Our results










may represent the first reporting that age-related decreases in bone mass occur in Swiss Webster

mice begin within a few months of these animals reaching peak bone mass. The effect was seen

in histomorphometry of the lumbar spine and both histomorphometry and microCT of the distal

femur.

There is evidence of variability between and within in-bred mouse strains with respect to

skeletal maturation (118,119) and, this study suggests there may be perhaps also age-related

changes to bone in young adult out-bred mice, like the Swiss Webster strain. One study found

that the Swiss Webster mouse reaches skeletal maturity between 5 and 6 months of age (5),

although no attempt was made to determine the age at which there would be a natural decline in

bone mass due to aging. Another study found that the level of osteoblastogenesis decreased

three-fold in SAMP6 mice between 3 and 4 months of age and resulted in significant bone loss

soon after (120). Data from another study using 10 week old C57BL/J6 mice showed a trend

toward decreasing bone volume at about 16 weeks of age (45). Our data also suggest an age-

related decrease in bone mass in control animals. The Swiss Webster mice in our study entered

at seven months of age and were sacrificed two months later. There was a lower cancellous

BV/TV in both the lumbar spine and distal femur, using microCT and histomorphometry,

between 7 month old and 9 month old control animals.

In the distal femur, both histomorphometry and microCT found 35-45% less BV/TV and

around a 20-40% lower Tb.N in nine month old animals than in their seven month old

counterparts. There was also more than double the Tb.Sp according to histomorphometry.

Despite these changes there was no significant difference in the Ob.S and Oc.S although Ob.S

showed a downward trend while Oc.S showed an upward trend. The downward trend in Ob.S

was insufficient to cause a significant difference in MS. However, there was a significant









decline in MAR, suggesting that, while osteoblast numbers may be relatively stable, the

osteoblasts were less actively forming bone. The 33.3% (p = 0.01) decline in MAR contributed

to the almost three-fold decline in BFR/B S in the 8 WK CNTL compared to the BSL CNTL

animals.

Prophylactic Value of Concurrent Treatment with Glucocorticoid Drugs and PTH

This study is the first to examine whether there is increased benefit to starting PTH

treatment concurrently with GC administration as opposed to using GC first and then later

treating with PTH. One group of animals was treated with GC for 4 weeks and then treated for

an additional four weeks with both GC and PTH. Another group simultaneously received GC

and PTH for 8 weeks. There was a benefit to simultaneous treatment but this may be the result

of the additional 4 weeks of PTH. There was a 47. 1% (p = 0.05) higher Ob.S and a 1 12.5% (p =

0.03) higher Oc. S with 8 weeks of PTH, indicating continued gains in the latter group. In the

lumbar vertebrae, the group treated with PTH for 8 weeks also had a significant decrease in

Tb.Sp. In the distal femur, significant changes were only seen with microCT, where both BV

and Tb.Th were significantly higher. There were no significant differences in MS, MAR, or

BFR/BS with additional PTH treatment. These findings suggest that even though 8 weeks of

PTH resulted in some additional improvement to bone parameters, 4 weeks of PTH after GC

exposure was still highly effective.

Comparing the effects of 4 weeks of PTH use against eight weeks of PTH in animals

receiving GC for 8 weeks shows that most of the change occurred in the first 4 weeks of PTH

administration. There were continued improvements in the group receiving PTH for 8 weeks,

but the magnitude of change was less. 82.9% of the final level of MS seen in the GC-PTH8

group was seen in the GC4/GC-PTH4 group. The same trend was true for MAR which was

slightly higher in the 4 week PTH group and BFR/BS where over 70% of the total 8 week










change was seen in the GC4/GC-PTH4 group. These findings suggest PTH was able to reverse

the effects of prior GC use sufficiently to make it a valid post-exposure treatment.

Site Specificity of GC and PTH Treatment

In mice, as in humans, BV/TV is higher in the vertebrae than in the femur. In humans

however, the effects of both GC and PTH are more prevalent in the vertebrae than in the femur.

Humans treated with both GC and PTH increased spinal BMD by 35% while the hip gained only

a modest 2% BMD over a one-year period (6). A different relationship has been reported in

mice (8) where there is greater change in long bones of the appendicular skeleton than in the

axial skeleton. Other studies in rats have reported the same results (39). The difference in

results found in rodents may result from biomechanical loading differences between bipeds and

quadrupeds (8). In our study, changes with both GC and PTH were also greater in the distal

femur than in the lumbar spine. Based on histomorphometry, BV/TV changes were greater in

the distal femur than in the lumbar spine (-46% versus -30%) and based on PTH exposure (100%

versus 42%). It may be that the biomechanics of weight-bearing activities of the mouse explain

the different findings or that there is an additive effect when PTH is acting on a bone already

exposed to higher mechanical loading (8, 121,122).

Conclusions

This study found that glucocorticoid drugs suppress bone turnover in the Swiss Webster

mouse and that these effects can be offset by PTH. The effects of PTH were rapid, with most of

the total changes seen over eight weeks occurring in the first 4 weeks. We also found that PTH

treatment after GC exposure is more effective at restoring bone mass than natural recovery. We

found there are site specific differences in response to PTH and that these caused greater changes

to occur in the appendicular rather than in the axial skeleton.









This study also reports some novel findings. We found there was an age-related loss of

bone between 7 and 9-month old animals who received only vehicle and no study drugs. We

also found that GCs tended to increase bone mass even though there was clearly a suppression of

bone turnover. Finally we found a residual effect with GC treatment that inhibited increases in

bone mass after GC treatment was discontinued even though osteoblastic activity rebounded

after GC removal.

Clinical Applications

The underlying goal for this research was to demonstrate the efficacy of using PTH to

mitigate the skeletal effects of GCs in mice, as a prelude to studies in humans. Swiss Webster

mice have previously been validated as models of bone loss due to ovariectomy (7) and for GC-

induced bone loss (2) and have been described as "a faithful model of the glucocorticoid-induced

bone loss in humans"(39). Despite the fact this study found exposure to GC may have increased

BV/TV, we believe the results pave the way for follow-on studies using PTH to treat bone loss

due to GCs in humans, although our results suggest the response of Swiss Webster mice to GCs

may be more variable than previously reported. There has been ample research demonstrating

the deleterious effects of GCs and the increased risk of bone fracture that accompanies their use

(21,23,73,76,81). This study demonstrated that PTH is effective in reversing the inhibitory

effects of GCs on bone formation. Our findings support a clinical application for PTH in

patients diseases conditions such as rheumatoid arthritis, COPD and other conditions that result

in long-standing GC use. PTH was used in one study with rheumatoid arthritis patients where it

increased spinal BMD by 35%, although gains in the hip were a more modest 2% (6). This

human study concluded that while anti-resorptive treatments can prevent bone loss, PTH was the

only therapy currently available that could reverse the suppressive effects of GCs (6). We

believe these findings and our current study show there is potential for successful treatment of









other patients who have experienced long-term glucocorticoid therapy including those that may

require solid organ transplants.

Due to the nature of post-transplant immunosuppressant protocols and reports of

osteosarcoma in rats exposed to life-time doses of PTH (103-105), we would not recommend the

use of PTH after transplant surgery, but this drug may prove efficacious in treating pre-existing,

GC-induced osteoporosis in those still awaiting surgery. The rapid effects of the PTH suggest

that even if patients could not complete the normal 18-24 month treatment regimen prior to

receiving a transplant, even a few months of treatment could be beneficial. Although beneficial

effects are more likely to be seen in the spine, one study showed that bone mass in the hip

increased six months after PTH was discontinued (44), indicating therapeutic benefits beyond the

dosing period.

Study Limitations

As with any study, resource considerations influenced study design and implementation.

Animals were housed one to a cage although this was not ideal. Mice are social animals that

interact with each other so housing them alone might have affected activity or stress levels. We

had to house the mice singly, however, to avoid Eighting. These were adult males formerly used

for breeding and were highly territorial. We also had to accept delivery of study animals in 7

shipments of 10 mice rather than receiving all animals simultaneously. This was necessary

because of the age of the animals involved. The other alternative would have been to buy young

animals and age them either at our facility or with the vendor. This would have decreased cohort

variation but was cost-prohibitive. We minimized the impact of multiple cohorts by assigning

animals from each arriving shipment to each study group.









Future Directions

There were novel findings in this study that warrant further investigation. The Einding of

negative changes in bone structural parameters in control mice over the two-month course of the

study was unexpected. Further studies are needed to verify these differences and establish the

significance and mechanisms mediating these changes. If there is a natural loss of bone so

quickly after attaining peak bone mass, researchers need to be aware of this so they don't

attribute age-related changes to interventional treatments.

The residual effect of GCs on the bone matrix also needs further study. Animals treated

with GCs and then allowed to recover showed further declines in bone structural parameters

despite increased osteoblastic activity. It is possible bone structural parameters would have

returned to normal if more recovery time had been allowed, but it is also possible that the GCs

caused a change to the bone matrix that inhibited recovery. If the latter is true, it might partially

explain the increased risk of bone fracture after GC therapy.

Finally, we believe the results of this study provide clear evidence of the efficacy of PTH

in treating bone changes caused by GC exposure. We believe human clinical populations may

also benefit from this treatment as previously discussed.












APPENDIX A
SUMMARY OF BONE MEASUREMENTS


A-1. Summary of Significant Changes in Lumbar Vertebrae L3 based on Histomorphometry.
BSL CNTL 8WK CNTL GC4/SAC GC4/RECV GC8 GC4/GC-PTH4 GC-PTH8
BSL CNTL BV/TV BV/TV BV/TV
-3).7% -27.7% -31.5
(p = ).()3) (p = .()3) (p = ).()2)


Tb.Wi
-18.4%
(p = ).()2)


Tb.Wi
-19.()
(p=).()2)


Tb.Wi
-25.1
(p = ().0)()


8 WK CNTL


BV/TV
+ 42.()%
(p = .()1)

Tb.Wi
+ 22.1
(p = .()1)

Tb.Sp
-24.7 %
(p = ).()3)


Tb.Wi
+ 24.5 %
(p =).()1)


GC4/ SAC


Tb.Wi
+ 14.5%
(p = .()3)


GC4/RECV


BV/TV
+35.9%
(p = .()3)

Tb.Wi
+23.1%
(p = .()1)


BV/TV
+ 43.7%
(p = ).()1)

Tb.Wi
+ 33.2%
( p= (.0)()


Tb.N
+ 19.5%
(p =).()2)

Tb.Sp
- 17.4%
(p = .()1)


Tb.Wi
+25.5%
(p = (.()))


BV/TV
+ 26.4%
(p = .()3)

Tb.Wi
+ 35.8%
(p = (.()))


GC 8


Tb.Wi
+ 18.5
(p = ).()2)


GC4/GC-PTH4












A-2. Summary of Significant Changes in Lumbar
GC4/RECV
BSL CNTL BV/TV
21.1%
(p =0.03)

Tb.N
9.5%
(p =0.04)


Vertebrae L2 based on MicroCT.
GC4/GC-PTH4 GC-PTH8
Tb.N Tb.Th
-9.5% + 12.3%
(p = 0.02) (p = 0.01)

Tb.Sp
+ 12.3%
( p= 0.04)


Tb.Th
-5.7%
(p =0.02)

Tb.Sp
+ 12.7%
(p =0.03)


8 WK CNTL


Tb.Th
+ 9.7%
(p = 0.04)


BV/TV
+ 36.6%
(p = 0.01)

Tb.Th
+ 15.2%
(p = 0.00)


BV/TV
+ 26.4%
(p = 0.04)

Tb.Th
+ 16.1%
(p = 0.00)


BV/TV
+ 46.7%
(p = 0.00)

Tb.Th
+ 19.2%
(p = 0.00)


Tb.Th
+ 17.9%
(p = 0.00)


Tb.Th
+ 10.6%
(p = 0.01)


GiC4/SAC


GC4/RECV


Tb.Th
+ 13.5%
(p = 0.00)


GC 8


BV/TV
-21.1%
(p =0.04)


Tb.Th
+ 12.3%
(p = 0.00)












of Significant
8WK CNTL
BV/TV
- 45.8%
(p =0.03)

Tb.N
- 40.0%
(p =0.05)


Changes in the Distal Femur based on Histomorphometry.
GC4/RECV GC8 GC4/GC-PTH4 GC-PTH8
BV/TV
52.1%
(p =0.01)

Tb.N
40.0%
(p =0.01)


A-3. Summary


BSL CNTL


Tb.Wi
-22.3%
(p =0.04)

Tb.Sp
+ 177.2%
(p =0.01)


Tb.Wi
-17.7%
(p =0.05)


Tb.Sp
+ 153.0%
(p =0.05)


8 WK CNTL


BV/TV
+ 100.0%
(p =0.02)

Tb.N
+ 83.3%
(p =0.03)

Tb.Sp
-59.4%
p =0.03


BV/TV
+ 126.1%
(p =0.01)

Tb.N
+ 83.3%
(p =0.03)

Tb.Wi
+ 30.6%
(p =0.02)

Tb.Sp
-63.0%
(p = 0.03)


Tb.Wi
+ 23.3%
P= 0.03


GC4/RECV


BV/TV
+ 95.6%
(p = 0.01)


Tb.Wi
+ 27.4%
(P =0.05)

Tb.Sp
-63.3%
(p = 0.01)


Tb.Wi
+ 20.3%
(p =0.04)


GC 8












of Significant
8WK CNTL
BV/TV
- 37.4%
(p = 0.00)

Tb.N
- 22.2%
( p=0.05)


Changes in the Distal Femur based on MicroCT
GC4/RECV GC8 GC4/GC-PTH4 GC-PTH8
BV/TV
39.4%
(p = 0.00)


A-4. Summary


BSL CNTL


Tb.Sp
+30.5%
(p = 0.05)


Tb.Th
+ 16.3%
(p = 0.01)

BV/TV
+ 85.5%
(p= 0.00)

Tb.N
+ 33.3%
(p = 0.00)


8 WK CNTL


BV/TV
+ 40.3%
(p =0.03)

Tb.N
+ 33.3%
(p = 0.03)

Tb.Th
- 12.7%
(p = 0.04)

Tb.Sp
- 19.4%
(p = 0.03)


Tb.Sp
- 25.6%
(p = 0.00)

Tb.Th
+ 17.5%
(p = 0.00)

BV/TV
+ 91.7%
(p= 0.00)

Tb.N
+ 27.3%
(p =0.01)

Tb.Th
+ 17.9%
(p = 0.00)

Tb.Sp
-27.8%
(p = 0.01)

BV/TV
+ 32.3%
p =0.04

Tb.Th
+25.4%
P =0.00
BV/TV
+ 45.6%
( p= 0.02)

Tb.Th
+ 10.5%
(p = 0.05)


GC4/SAC



GC4/RECV


Tb.Th
+ 6.3%
( p= 0.05)


BV/TV
+ 45.0%
(p=0.02)


GC 8


Tb.Th
+ 13.5%
p =0.05


GC4/
GC-PTH4












A-5. Percent Changes in Osteoclast and Osteoblast Surfaces in the Distal Femur.
p GC4/SAC GC4/RECV GC8 GC4/GC-PTH4 GC-PTH8
BSL CNTL Oc.S Oc.S
+ 116.7% + 183.3%
(p = 0.02) (p = 0.01)

Ob.S
+ 153.2%
(p = 0.01)


8 WK CNTL


Oc.S
-63.6%
(p = 0.03)

Ob.S
- 84.7%
(P= 0.00)


Ob.S
+ 177.8%
(p =0.00)



Oc.S
+ 325.0%
(p = 0. 00)

Ob.S
+ 1,718.2%
(p = 0.00)



Ob.S
+ 129.9%
(p = 0.00)



Ob.S
+ 852.4%
( p= 0.00)



Oc.S
+ 112.5%
(p = 0.03)

Ob.S
+47.1%
(p = 0.05)


Ob.S
-70.8
(p = 0.00)


GC4/SAC


Oc.S
+ 225%
(p = 0.00)

Ob.S
+ 690.9%
(p = 0.00)


Ob.S
+1,136.4%
(p= 0.00)


GC4/RECV


GC 8


Ob.S
+314.3%
(p = 0. 00)


Ob.S
+ 547.6
(p = 0.00)


GC4/GC-PTH4












A-6. Percent Changes in Dynamic Bone Formation Parameters in the Distal Femur.
8WK CNTL GC4/SAC GC4/RECV GC8 GC4/GC-PTH4 GC-PTH8
BSL CNTL MS MS MS MS
83.8% 76.8% +111.9% + 150.7%
(p = 0.01) (p = 0.02) (p =0.01) (p = 0.00)


MAR MAR MAR
-33.3% 33.3% 33.3%
(p = 0.01) (p = 0.01) (p =0.02)

BFR/BS BFR/BS
84.4% + 131.3%
(p =0.01) (p = 0.02)

8 WK CNTL MS MS MS MS MS
64.5% + 90.3% 48.4% + 358.1% + 441.9%
(p = 0.02 (p = 0.05) (p =0.05) ( p= 0.00) (p = 0.00)

MAR MAR MAR
+ 16.7% + 33.3% + 33.3%
(p = 0.03) (p =0.01) (p = 0.02)

BFR/BS BFR/BS BFR/BS BFR/BS
-72.7% + 90.9% + 390.9% + 572.7%
(p = 0.02)) (p = 0.03) (p =0.01) ( p= 0.00)

GC4/SAC MS MS MS
+ 436.4% + 1,190.9% + 1,427.3%
(p = 0.00) (p =0.00) (p = 0.00)

MAR MAR MAR
+ 16.7% +33.3% + 33.3%
(p = 0.03) (p =0.01) (p = 0.02)

BFR/BS BFR/BS BFR/BS
+ 600.0% + 1,700.0% + 2,366.7%
(p = 0.00) (p =0.00) (p = 0.00)

MS MS
GC4/RECV + 140.7% + 184.7%
(p =0-.01) (p = 0.00)

MAR
+ 14.3%
(p =0.02)

BFR/BS BFR/BS
+ 157.1% + 254.4%
(p =0.02) (p = 0.00)

GC 8 MS MS MS
+268.8% +787.5% + 950.0%
(p = 0.00) (p =0.00) (p = 0.00)

MAR
+ 33.3%
(p =0.02)

BFR/BS BFR/BS BFR/BS
+76.2% + 980.0% + 1380.0%
(p = 0.00) (p =0.00) (p = 0.00)









APPENDIX B
SUMMARY OF SELECTED STUDIES IN MICE












Table B-1. Studies Of Glucocorticoid-Induced Bone Loss In Mice.
Strain Treatment Dose Age Gender #/Grp
(mg/kg) (wk.)
SW Prednisolone 2.1 28 M 4-5
(pellet)


Time
(days)
27


Results

- Spinal BMD
decrease
- Preferential loss
in axial skeleton
- Increased
resorption
Decreased
formation
- Increased
osteocyte
apoptosis

- Decreased
Osteoclast
apoptosis
- Increased
osteoclast
survival
- Decreased bone
formation rate
- Increased
osteoblast
apoptosis


Type
Analysis
DXA

Hi stomor-
phometry


Weinstein
(5)

1998


Prednisolone (pellet) 2.1
Prednisolone +
alendronate


16 M


4, 10,
and 27


Weinstein
(4)


2002


DXA

Hi stomor-
phometry










Table B-1. Continued.
Strain Treatment


Dose
(mg/kg)
1 and 10


Age
_(wk.)_
28


Gender #/grp Time
(days)
F 5 21


Results


Type
Analysis
Hi stomor-
phometry

MicroCT


Balb/C Dexa-methasone (IP)


- Changes only
seen at higher
dosages

-Decreased
BFR/BS and
MAR


McLaughlin
(3)

2002


Biochemical
Assays


- Decreased
osteocalcin


Prednisolone (pellets) 1.4


24 M


Unk 21 -Decrease Tr bone Lane
volume and (2)
strength
2006
-Increased size of
osteocyte lacunae

-Increased DPD
crosslinks


Hi stomor-
phometry

MicroCT

Biochemical
Assays


resorptionn)

-Decreased
Osteocalcin
(formation)
BFR/BS = bone formation rate/bone surface; BMD = bone mineral density; DPD = deoxipyrodinoline; GC = glucocorticoid; Tr = trabecular; MAR = mineral
apposition rate; SW = Swiss Webster.










Table B-2. Studies using Teriparatide in Mice.
Strain Condition Dose Age Gender N/ Time Results Type
(Clg/kg) (wk.) grp (Wks) Analysis


C57BL/6 Ovx/sham 40 12


4-6 3 or 7 Increased bone
formation rate and
BV/TV
Increased Oc. S
-greater effect in
LV than tibia
Little change in
cortical bone
9 3 and 7 -greater BMD
change in tibia and
femur than LV
increased bone
turnover
9 4 -Bone loss reversed
Mice lost bone
faster than rats
2-3X increase


Hi stomor-
phometry


Zhou (8)


C57BL/6 Intact


10 F





11 F


lida-Klein
(45)



Alexander
(7)


Piximus
(DXA)

Hi stomor-
phometry
Hi stomor-
phometry

MicroCT


Ovx


MAR
BMD = bone mineral density; BV/TV = bone volume/total volume; LV = lumbar vertebrae; MAR = mineral apposition rate; Oc. S = osteoclast surface; Ovx
ovariectomized; SW = Swiss Webster.









LIST OF REFERENCES


1. Hofbauer LC, Gori F, Riggs BL, Lacey DL, Dunstan CR, Spelsberg TC, Khosla S 1999
Stimulation of osteoprotegerin ligand and inhibition of osteoprotegerin production by
glucocorticoids in human osteoblastic lineage cells: potential paracrine mechanisms of
glucocorticoid-induced osteoporosis. Endocrinology 140(10):4382-9.

2. lane NE, Yao W, Balooch M, Nalla R, Balooch G, Habelitz S, Kinney J, Bonewald LF
2006 Glucocorticoid-treated mice have localized changes in trabecular bone material
properties and osteocyte lacunar size that are not observed in placebo-treated or estrogen-
deficient mice. Journal of Bone and Mineral Research 21(14):466-476.

3. McLaughlin F, Mackintosh J, Hayes BP, McLaren A, Uings IJ, Salmon P, Humphreys J,
Meldrum E, Farrow SN 2002 Glucocorticoid-induced osteopenia in the mouse as
assessed by histomorphometry, microcomputed tomography, and biochemical markers.
Bone 30(6):924-30.

4. Weinstein RS, Chen JR, Powers CC, Stewart SA, Landes RD, Bellido T, Jilka RL, Parfitt
AM, Manolagas SC 2002 Promotion of osteoclast survival and antagonism of
bisphosphonate-induced osteoclast apoptosis by glucocorticoids. J Clin Invest
109(8):1041-8.

5. Weinstein RS, Jilka RL, Parfitt AM, Manolagas SC 1998 Inhibition of osteoblastogenesis
and promotion of apoptosis of osteoblasts and osteocytes by glucocorticoids. Potential
mechanisms of their deleterious effects on bone. J Clin Invest 102(2):274-82.

6. Lane NE, Sanchez S, Modin GW, Genant HK, Pierini E, Arnaud CD 1998 Parathyroid
hormone treatment can reverse corticosteroid-induced osteoporosis. Results of a
randomized controlled clinical trial. J Clin Invest 102(8): 1627-3 3.

7. Alexander JM, Bab I, Fish S, Muller R, Uchiyama T, Gronowicz G, Nahounou M, Zhao
Q, White DW, Chorev M, Gazit D, Rosenblatt M 2001 Human parathyroid hormone 1-34
reverses bone loss in ovariectomized mice. J Bone Miner Res 16(9): 1665-73.

8. Zhou H, lida-Klein A, Lu SS, Ducayen-Knowles M, Levine LR, Dempster DW, Lindsay
R 2003 Anabolic action of parathyroid hormone on cortical and cancellous bone differs
between axial and appendicular skeletal sites in mice. Bone 32(5):513-20.

9. Braith RW, Magyari PM, Fulton MN, Aranda J, Walker T, Hill JA 2003 Resistance
exercise training and alendronate reverse glucocorticoid-induced osteoporosis in heart
transplant recipients. J Heart Lung Transplant 22(10): 1082-90.

10. Braith RW, Mills RM, Welsch MA, Keller JW, Pollock ML 1996 Resistance exercise
training restores bone mineral density in heart transplant recipients. J Am Coll Cardiol
28(6):1471-7.

11. Boulos P, loannidis G, Adachi JD 2000 Glucocorticoid-induced osteoporosis. Curr
Rheumatol Rep 2(1):53-61.










12. Clarke B, Leidig-Bruckner G 2005 Fracture Prevalence and Incidence in Solid Organ
Transplant Recipients. In: Compston J, Shane E (eds.) Bone Disease of Organ
Transplantation. El Sevier Academic Press, Boston.

13. Dennison E, Cooper C 2002 Epidemiology of glucocorticoid-induced osteoporosis. Front
Horm Res 30:121-6.

14. Rehman Q, Lane NE 2003 Effect of glucocorticoids on bone density. Med Pediatr Oncol
41(3):212-6.

15. Sambrook P, Lane NE 2001 Corticosteroid osteoporosis. Best Pract Res Clin Rheumatol
15(3):401-13.

16. Boling EP 2004 Secondary osteoporosis: underlying disease and the risk for
glucocorticoid-induced osteoporosis. Clin Ther 26(1): 1-14.

17. Cohen A, Shane E 2003 Osteoporosis after solid organ and bone marrow transplantation.
Osteoporos Int 14(8):617-30.

18. Cohen D, Adachi JD 2004 The treatment of glucocorticoid-induced osteoporosis. J
Steroid Biochem Mol Biol 88(4-5):337-49.

19. Rodino MA, Shane E 1998 Osteoporosis after organ transplantation. Am J Med
104(5):459-69.

20. Shane E, Epstein S 2001 Transplantation Osteoporosis. Transplantation Reviews


21. Dalle Carbonare L, Arlot ME, Chavassieux PM, Roux JP, Portero NR, Meunier PJ 2001
Comparison of trabecular bone microarchitecture and remodeling in glucocorticoid-
induced and postmenopausal osteoporosis. J Bone Miner Res 16(1):97-103.

22. McIlwain HH 2003 Glucocorticoid-induced osteoporosis: pathogenesis, diagnosis, and
management. Prev Med 36(2):243-9.

23. Lafage-Proust MH, Boudignon B, Thomas T 2003 Glucocorticoid-induced osteoporosis:
pathophysiological data and recent treatments. Joint Bone Spine 70(2): 109-18.

24. Braith RW, Magyari PM, Fulton MN, Lisor CF, Vogel SE, Hill JA, Aranda JM
Comparison of calcitonin versus calcitonin + resistance exercise as prophylaxis for
osteoporosis in heart transplant recipients. Transplantation In Press.

25. Braith RW, S.D. G, Musto T, Mitchell MJ, Baz MA 1998 Resistance exercise restored
bone mineral density in an osteoporotic patient before lung transplantation. Journal of
Cardiopulmonary Rehabilitation 36: 18-23.









26. Mitchell M, Fulton M, Lisor C, Baz M, Braith R 2002 Resistance training attenuates
glucocorticoid-induced osteoporosis in lung transplant recipients. Journal of Heart and
Lung Transplantation in review.

27. Uusi-Rasi K, Sievanen H, Heinonen A, Kannus P, Vuori I 2004 Effect of discontinuation
of alendronate treatment and exercise on bone mass and physical fitness: 15-month
follow-up of a randomized, controlled trial. Bone 35(3):799-805.

28. Mitlak BH 2002 Parathyroid hormone as a therapeutic agent. Curr Opin Pharmacol
2(6):694-9.

29. Neer RM, Arnaud CD, Zanchetta JR, Prince R, Gaich GA, Reginster JY, Hodsman AB,
Eriksen EF, Ish-Shalom S, Genant HK, Wang O, Mitlak BH 2001 Effect of parathyroid
hormone (1-34) on fractures and bone mineral density in postmenopausal women with
osteoporosis. N Engl J Med 344(19): 1434-41.

30. Rittmaster RS, Bolognese M, Ettinger MP, Hanley DA, Hodsman AB, Kendler DL,
Rosen CJ 2000 Enhancement of bone mass in osteoporotic women with parathyroid
hormone followed by alendronate. J Clin Endocrinol Metab 85(6):2129-34.

31. Kaufman JM, Orwoll E, Goemaere S, San Martin J, Hossain A, Dalsky GP, Lindsay R,
Mitlak BH 2005 Teriparatide effects on vertebral fractures and bone mineral density in
men with osteoporosis: treatment and discontinuation of therapy. Osteoporos Int
16(5):510-6.

32. Finkelstein JS, Hayes A, Hunzelman JL, Wyland JJ, Lee H, Neer RM 2003 The effects of
parathyroid hormone, alendronate, or both in men with osteoporosis. N Engl J Med
349(13):1216-26.

33. Body JJ, Gaich GA, Scheele WH, Kulkarni PM, Miller PD, Peretz A, Dore RK, Correa-
Rotter R, Papaioannou A, Cumming DC, Hodsman AB 2002 A randomized double-blind
trial to compare the efficacy of teriparatide [recombinant human parathyroid hormone (1-
34)] with alendronate in postmenopausal women with osteoporosis. J Clin Endocrinol
Metab 87(10):4528-35.

34. Lindsay R, Nieves J, Formica C, Henneman E, Woelfert L, Shen V, Dempster D,
Cosman F 1997 Randomised controlled study of effect of parathyroid hormone on
vertebral-bone mass and fracture incidence among postmenopausal women on oestrogen
with osteoporosis. Lancet 350(9077):550-5.

35. Pettway GJ, Schneider A, Koh AJ, Widjaja E, Morris MD, Meganck JA, Goldstein SA,
McCauley LK 2005 Anabolic actions of PTH (1-34): use of a novel tissue engineering
model to investigate temporal effects on bone. Bone 36(6):959-70.

36. Binz K, Schmid C, Bouillon R, Froesch ER, Jurgensen K, Hunziker EB 1994 Interactions
of insulin-like growth factor I with dexamethasone on trabecular bone density and
mineral metabolism in rats. Eur J Endocrinol 130(4):387-93.









37. King CS, Weir EC, Gundberg CW, Fox J, Insogna KL 1996 Effects of continuous
glucocorticoid infusion on bone metabolism in the rat. Calcif Tissue Int 59(3): 184-91.

38. Shen V, Birchman R, Liang XG, Wu DD, Lindsay R, Dempster DW 1997 Prednisolone
alone, or in combination with estrogen or dietary calcium deficiency or immobilization,
inhibits bone formation but does not induce bone loss in mature rats. Bone 21(4):345-51.

39. Manolagas SC, Weinstein RS 1999 New developments in the pathogenesis and treatment
of steroid-induced osteoporosis. J Bone Miner Res 14(7): 1061-6.

40. Mosekilde L 1995 Assessing bone quality--animal models in preclinical osteoporosis
research. Bone 17(4 Suppl):343S-352S.

41. Thompson DD, Simmons HA, Pirie CM, Ke HZ 1995 FDA Guidelines and animal
models for osteoporosis. Bone 17(4 Suppl):125S-133S.

42. Weinstein RS 2001 Glucocorticoid-induced osteoporosis. Rev Endocr Metab Disord
2(1):65-73.

43. Weinstein RS, Manolagas SC 2000 Apoptosis and osteoporosis. Am J Med 108(2):153-
64.

44. Lane NE, Sanchez S, Modin GW, Genant HK, Pierini E, Arnaud CD 2000 Bone mass
continues to increase at the hip after parathyroid hormone treatment is discontinued in
glucocorticoid-induced osteoporosis: results of a randomized controlled clinical trial. J
Bone Miner Res 15(5):944-51.

45. lida-Klein A, Zhou H, lu SS, levine LR, Ducayen-Knowles M, Dempster D, Nieves J,
lindsay R 2002 Anabolic action of parathyroid hormone is skeletal site specific at the
tissue and cellular levels in mice. Journal of Bone and Mineral Research 17(5):808-816.

46. ALA July 2005 Estimated prevalence and incidence of lung disease by lung association
territory American Lung Association: Epidemiology and statistical unit research and
program services, pp 1-53.

47. 2005 Organ Procurement andTransplantation Network Statistical Database.

48. Swarthout JT, D'Alonzo RC, Selvamurugan N, Partridge NC 2002 Parathyroid hormone-
dependent signaling pathways regulating genes in bone cells. Gene 282(1-2): 1-17.

49. Wronski TJ, Yen C-F, Qi H, Dann LM 1993 Parathyroid hormone is more effective than
estrogen or bisphosphonates for restoration of lost bone mass in ovariectomized rats.
Endocrinology 132(2): 823-831.

50. Li M, Shen Y, Halloran BP, Baumann BD, Miller K, Wronski TJ 1996 Skeletal response
to corticosteroid deficiency and excess in growing male rats. Bone 19(2):81-8.









51. Parfitt AM, Drezner MK, Glorieux FH, Kanis JA, Malluche H, Meunier PJ, Ott SM,
Recker RR 1987 Bone histomorphometry: standardization of nomenclature, symbols, and
units. Report of the ASBMR Histomorphometry Nomenclature Committee. J Bone Miner
Res 2(6):595-610.

52. Conover WJ 1980 Practical Nonparametric Statistics. Wiley and Sons, New York, NY,
pp 229-237.

53. Bonner F, Worrell RV 1991 A Basic Science Primer in Orthopaedics.

54. Reeve J 2000 How do women develop fragile bones? J Steroid Biochem Mol Biol
74(5):375-81.

55. Kanis J 1991 Calcium Requirements for Optimal Skeletal Health in Women. Calcified
Tissue International Supplement 49:S33-S41.

56. Parfitt AM 1992 Implications of Architecture for the Pathogenesis and Prevention of
Vertebral Fracture. bone 13(Supplement): S41-S47.

57. Vaananen HK, Zhao H, Mulari M, Halleen JM 2000 The Cell Biology of Osteoclast
Function. J Cell Sci 113:377-381.

58. Arita S, Ikeda S, Sakai A, Okimoto N, Akahoshi S, Nagashima M, Nishida A, Ito M,
Nakamura T 2004 Human parathyroid hormone (1-34) increases mass and structure of
the cortical shell, with resultant increase in lumbar bone strength, in ovariectomized rats.
J Bone Miner Metab 22(6):530-40.

59. Vaananen HK, Zhao H, Mulari M, Halleen JM 2000 The cell biology of osteoclast
function. J Cell Sci 113 ( Pt 3):377-81.

60. Hofbauer LC, Khosla S, Dunstan CR, Lacey DL, Boyle WJ, Riggs BL 2000 The roles of
osteoprotegerin and osteoprotegerin ligand in the paracrine regulation of bone resorption.
J Bone Miner Res 15(1):2-12.

61. Martin RB 2000 Toward a Unifying Theory of Bone Remodeling. Bone 26(1).

62. Martin RB, Burr DB, Sharkey NA 1998 Skeletal Tissue Mechanics. Springer-Verelag,
New York.

63. Notelovitz M, Martin D, Tesar R 1991 Estrogen Therapy and Variable-Resistance weight
training increase bone mineral in surgically menopausal women. Journal of Bone and
Mineral Research 6:583-590.

64. Hazelwood SJ, Bruce Martin R, Rashid MM, Rodrigo JJ 2001 A mechanistic model for
internal bone remodeling exhibits different dynamic responses in disuse and overload. J
Biomech 34(3):299-308.









65. Takahashi N, Udagawa N, Suda T 1999 A new member of tumor necrosis factor ligand
family, ODF/OPGL/TRANCE/RANKL, regulates osteoclast differentiation and function.
Biochem Biophys Res Commun 256(3):449-55.

66. Frost H 1997 On Our Age-Related Bone Loss: Insights from a New Paradigm. Journal of
Bone and Mineral Research 12(10).

67. Bronner F, Worrell RV 1991 A Basic Science Primer in Orthopaedics. Lippincott,
WIlliams, and Wilkins, New York.

68. Hazelwood SJ, Martin RB, Rashid MM, Rodrigo J 2001 A Mechanistic Model for
Internal Bone Remodeling Exhibits Different Dynamic Responses in Disuse and
Overload. Journal of Biomechanics 34:299-308.

69. Kostenuik PJ 2005 Osteoprotegerin and RANKL regulate bone resorption, density,
geometry and strength. Curr Opin Pharmacol 5(6):618-25.

70. Oh ES, Rhee EJ, Oh KW, Lee WY, Baek KH, Yoon KH, Kang MI, Yun EJ, Park CY,
Choi MG, Yoo HJ, Park SW 2005 Circulating osteoprotegerin levels are associated with
age, waist-to-hip ratio, serum total cholesterol, and low-density lipoprotein cholesterol
levels in healthy Korean women. Metabolism 54(1):49-54.

71. Sandy J, Davies M, Prime S, Farndale R 1998 Signal pathways that transduce growth
factor-stimulated mitogenesis in bone cells. Bone 23(1): 17-26.

72. Dalle Carbonare L, Bertoldo F, Valenti MT, Zenari S, Zanatta M, Sella S, Giannini S,
Cascio VL 2005 Histomorphometric analysis of glucocorticoid-induced osteoporosis.
Micron 36(7-8):645-52.

73. Dalle Carbonare L, Chavassieux PM, Arlot ME, Meunier PJ 2002 Bone
histomorphometry in untreated and treated glucocorticoid-induced osteoporosis. Front
Horm Res 30:37-48.

74. Patschan D, Loddenkemper K, Buttgereit F 2001 Molecular mechanisms of
glucocorticoid-induced osteoporosis. Bone 29(6):498-505.

75. Reid IR 2000 Glucocorticoid-induced osteoporosis. Baillieres Best Pract Res Clin
Endocrinol Metab 14(2):279-98.

76. Manelli F, Giustina A 2000 Glucocorticoid-induced osteoporosis. Trends Endocrinol
Metab 11(3):79-85.

77. Adcock IM 2004 Corticosteroids: limitations and future prospects for treatment of severe
inflammatory disease. Drug Discovery Today: Therapeutic Strategies 1(3):321-328.

78. Demoly P, Chung KF 1998 Pharmacology of corticosteroids. Respir Med 92(3):385-94.









79. Schacke H, Docke WD, Asadullah K 2002 Mechanisms involved in the side effects of
glucocorticoids. Pharmacol Ther 96(1):23-43.

80. Canalis E 2003 Mechanisms of glucocorticoid-induced osteoporosis. Curr Opin
Rheumatol 15(4):454-7.

81. Canalis E, Bilezikian JP, Angeli A, Giustina A 2004 Perspectives on glucocorticoid-
induced osteoporosis. Bone 34(4):593-8.

82. Ton FN, Gunawardene SC, Lee H, Neer RM 2005 Effects of low-dose prednisone on
bone metabolism. J Bone Miner Res 20(3):464-70.

83. Boyde A, Maconnachie E, Reid SA, Delling G, Mundy GR 1986 Scanning electron
microscopy in bone pathology: review of methods, potential and applications. Scan
Electron Microsc (Pt 4): 153 7-54.

84. Tamura Y, Okinaga H, Takami H 2004 Glucocorticoid-induced osteoporosis. Biomed
Pharmacother 58(9):500-4.

85. Malyszko J, Malyszko JS, Wolczynski S, Mysliwiec M 2003 Osteoprotegerin and its
correlations with new markers of bone formation and bone resorption in kidney transplant
recipients. Transplant Proc 35(6):2227-9.

86. Ferrari P 2003 Cortisol and the renal handling of electrolytes: role in glucocorticoid-
induced hypertension and bone disease. Best Pract Res Clin Endocrinol Metab 17(4):575-
89.

87. Canalis E 1996 Clinical review 83: Mechanisms of glucocorticoid action in bone:
implications to glucocorticoid-induced osteoporosis. J Clin Endocrinol Metab
81(10):3441-7.

88. Sivagurunathan S, Muir MM, Brennan TC, Seale JP, Mason RS 2005 Influence of
glucocorticoids on human osteoclast generation and activity. J Bone Miner Res
20(3):390-8.

89. Reid DM, Harvie J 1997 Secondary osteoporosis. Baillieres Clin Endocrinol Metab
11(1):83-99.

90. Brixen KT, Christensen PM, Ej ersted C, Langdahl BL 2004 Teriparatide (biosynthetic
human parathyroid hormone 1-34): a new paradigm in the treatment of osteoporosis.
Basic Clin Pharmacol Toxicol 94(6):260-70.

91. Fox J 2002 Developments in parathyroid hormone and related peptides as bone-formation
agents. Curr Opin Pharmacol 2(3):338-44.

92. Debiais F 2003 Efficacy data on teriparatide (parathyroid hormone) in patients with
postmenopausal osteoporosis. Joint Bone Spine 70(6):465-70.









93. Selye H 1932 On the stimulation of new bone formation with parathyroid extract and
irradiated ergosterol. Endocrinology 16:547-558.

94. Fitzpatrick LA, Bilezikian JP 1996 Actions of Parathyroid Hormone. In: Bilezikian JP,
Raisz LG, Rodan GA (eds.) Principles of Bone Biology. Academic Press, San Diego.

95. Locklin RM, Khosla S, Turner RT, Riggs BL 2003 Mediators of the biphasic responses
of bone to intermittent and continuously administered parathyroid hormone. J Cell
Biochem 89(1):180-90.

96. Canalis E, Centrella M, Burch W, McCarthy TL 1989 Insulin-like growth factor I
mediates selective anabolic effects of parathyroid hormone in bone cultures. J Clin Invest
83(1):60-5.

97. Rosen CJ 2004 What's new with PTH in osteoporosis: where are we and where are we
headed? Trends Endocrinol Metab 15(5):229-33.

98. Gensure RC, Gardella TJ, Juppner H 2005 Parathyroid hormone and parathyroid
hormone-related peptide, and their receptors. Biochem Biophys Res Commun
328(3):666-78.

99. Rosen CJ, Bilezikian JP 2001 Clinical review 123: Anabolic therapy for osteoporosis. J
Clin Endocrinol Metab 86(3):957-64.

100. Quattrocchi E, Kourlas H 2004 Teriparatide: a review. Clin Ther 26(6):841-54.

101. Onyia JE, Helvering LM, Gelbert L, Wei T, Huang S, Chen P, Dow ER, Maran A, Zhang
M, Lotinun S, Lin X, Halladay DL, Miles RR, Kulkarni NH, Ambrose EM, Ma YL,
Frolik CA, Sato M, Bryant HU, Turner RT 2005 Molecular profile of catabolic versus
anabolic treatment regimens of parathyroid hormone (PTH) in rat bone: an analysis by
DNA microarray. J Cell Biochem 95(2):403-18.

102. Xing L, Boyce BF 2005 Regulation of apoptosis in osteoclasts and osteoblastic cells.
Biochem Biophys Res Commun 328(3):709-20.

103. Tashjian AH, Jr., Chabner BA 2002 Commentary on clinical safety of recombinant
human parathyroid hormone 1-34 in the treatment of osteoporosis in men and
postmenopausal women. J Bone Miner Res 17(7): 1151-61.

104. Vahle JL, Long GG, Sandusky G, Westmore M, Ma YL, Sato M 2004 Bone neoplasms in
F344 rats given teriparatide [rhPTH(1-34)] are dependent on duration of treatment and
dose. Toxicol Pathol 32(4):426-38.

105. Vahle JL, Sato M, Long GG, Young JK, Francis PC, Engelhardt JA, Westmore MS,
Linda Y, Nold JB 2002 Skeletal changes in rats given daily subcutaneous inj sections of
recombinant human parathyroid hormone (1-34) for 2 years and relevance to human
safety. Toxicol Pathol 30(3):312-21.









106. Berg C, Neumeyer K, Kirkpatrick P 2003 Teriparatide. Nat Rev Drug Discov 2(4):257-8.

107. U.S.Forteo 2004 Prescribing information, Eli Lilly and Company.

108. Qin L, Raggatt LJ, Partridge NC 2004 Parathyroid hormone: a double-edged sword for
bone metabolism. Trends Endocrinol Metab 15(2):60-5.

109. Audran M, Insalaco P 2003 Parathyroid hormone therapy for osteoporosis. Joint Bone
Spine 70(5):315-7.

110. Jiang Y, Zhao JJ, Mitlak BH, Wang O, Genant HK, Eriksen EF 2003 Recombinant
human parathyroid hormone (1-34) [teriparatide] improves both cortical and cancellous
bone structure. J Bone Miner Res 18(11): 1932-41.

111. Dempster DW, Cosman F, Kurland ES, Zhou H, Nieves J, Woelfert L, Shane E, Plavetic
K, Muller R, Bilezikian J, Lindsay R 2001 Effects of daily treatment with parathyroid
hormone on bone microarchitecture and turnover in patients with osteoporosis: a paired
biopsy study. J Bone Miner Res 16(10): 1846-53.

112. Adams AE, Rosenblatt M, Suva LJ 1999 Identification of a novel parathyroid hormone-
responsive gene in human osteoblastic cells. Bone 24(4):305-13.

113. Samuels A, Perry MJ, Gibson RL, Colley S, Tobias JH 2001 Role of endothelial nitric
oxide synthase in estrogen-induced osteogenesis. Bone 29(1):24-9.

114. Weinstein RS, Jia D, Powers CC, Stewart SA, Jilka RL, Parfitt AM, Manolagas SC 2004
The skeletal effects of glucocorticoid excess override those of orchidectomy in mice.
Endocrinology 145(4): 1980-7.

115. Oxlund H, Ortoft G, Thomsen JS, Danielsen CC, Ej ersted C, Andreassen TT 2006 The
anabolic effect of PTH on bone is attenuated by simultaneous glucocorticoid treatment.
Bone 39(2):244-52.

116. Samuels A, Perry MJ, Gibson R, Tobias JH 2001 Effects of combination therapy with
PTH and 17beta-estradiol on long bones of female mice. Calcif Tissue Int 69(3): 164-70.

117. Sibonga JD, Iwaniec UT, Shogren KL, Rosen CJ, Turner RT 2007 Effects of parathyroid
hormone (1-34) on tibia in an adult rat model for chronic alcohol abuse. Bone
40(4):1013-20.

118. Knopp E, Troiano N, Bouxsein M, Sun BH, Lostritto K, Gundberg C, Dziura J, Insogna
K 2005 The effect of aging on the skeletal response to intermittent treatment with
parathyroid hormone. Endocrinology 146(4):1983-90.

119. Halloran BP, Ferguson VL, Simske SJ, Burghardt A, Venton LL, Majumdar S 2002
Changes in bone structure and mass with advancing age in the male C57BL/6J mouse. J
Bone Miner Res 17(6):1044-50.









120. Jilka RL, Weinstein RS, Takahashi K, Parfitt AM, Manolagas SC 1996 Linkage of
decreased bone mass with impaired osteoblastogenesis in a murine model of accelerated
senescence. J Clin Invest 97(7): 1732-40.

121. Turner RT, Evans GL, Lotinun S, Lapke PD, Iwaniec UT, Morey-Holton E 2007 Dose-
response effects of intermittent PTH on cancellous bone in hindlimb unloaded rats. J
Bone Miner Res 22(1):64-71.

122. Turner RT, Lotinun S, Hefferan TE, Morey-Holton E 2006 Disuse in adult male rats
attenuates the bone anabolic response to a therapeutic dose of parathyroid hormone. J
Appl Physiol 101(3):881-6.










BIOGRAPHICAL SKETCH

Kathleen S. Howe was commissioned as an officer in the United States Air Force after her

graduation from the University of Central Florida in 1978. She spent the next 20 years serving in

a number of positions in the United States Air Force before retiring as a Lieutenant Colonel in

2000. After her retirement, she returned to the academic world to pursue her interest in bone

metabolism. She has spent the past 7 years concentrating on her studies in exercise physiology

and bone biology. Her research has centered on interventions to reverse the effects of aging and

disease on bone and reversing the effects of secondary osteoporosis. She plans to continue

research related to bone metabolism in the future. She has accepted a postdoctoral position in

the Department of Nutrition and Exercise Science at Oregon State University where she will

examine the effects of alcohol on bone structure and metabolism.





PAGE 1

1 SKELETAL EFFECTS OF TER IPARATIDE IN GLUCOCORTICOID-TREATED MICE By KATHLEEN S. HOWE A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2007

PAGE 2

2 2007 Kathleen S. Howe

PAGE 3

3 To my mother and father. My fathers recent passing brought great sadness to our family, but I know he would be proud of this accomplishment. My mother has been extremely supportive and her encouragement was instrumental in making this dissertation a reality.

PAGE 4

4 ACKNOWLEDGMENTS No project of this size occurs without a grea t deal of support and encouragement. First, and most importantly, I wish to thank the many individuals who directly contributed to this research study. Jodi Long was involved in ev ery aspect of this study. Her enthusiasm, knowledge, and energy contributed immensely to this pr oject. I also owe a debt of gratitude to Dr. Randy Braith for his guidance and support th roughout this project and my tenure at this university. His extensive knowle dge and support were invaluable to me as I designed and conducted this study. His comments, suggestions, and editorial talents helped bring the study results into sharper focus. I owe a particular debt of gratitude to Dr. Tom Wronski who helped design this study, allowed me to use equipment in his lab, and helped me to interpret the results. He is the foremost expert on bone biology at the University of Florida and I prof ited greatly from his knowledge. I would also like to thank Molly Altman and Sally Vanegas who prepared and analyzed bone samples for histomorphometric analysis and Ignacio A guirre who was always willing to answer my questions. Their knowledge and professionalism helped make this project a success. I would also like to thank Dr. Russell Turner and Dr. Urszula Iwaniec. They graciously permitted me access to their equipment and expertis e. Their enthusiasm for this project and willingness to teach me about bone research has helped my professional development immensely. I have also benefited greatly fr om my committees guida nce and thank both Dr. Stephan Dodd and Dr. Scott Powers. Their insigh tful comments and probing questions helped make this a better study. More than that, though, over the past seven years they have provided encouragement and helped shape my professional deve lopment. I also owe a debt of gratitude to the wonderful staff at Animal Care Services. They provided outstanding support, training, and

PAGE 5

5 services to this project. Working with them wa s truly a pleasure. I w ould also like to thank some of my friends who supported me: Susan Smith who always had a kind word, Vija Purs who grudgingly accepted my research with animals, Ben Webster, who kept me smiling, and, most of all Janet Degner who was always there with a word of encouragement when the going got rough. I also want to thank members of the Applied Physiology and Kinesiology staff. Kim Hatch and Candyce Hudson have provided expert advice and support over the years. Their willingness to help students and their expertise in managing grants greatly eased the way for this project. I would also like to thank James Milford and Susie Weldon. They are the collective memory for the department and seem to always know how to make thin gs happen. Their cando attitude and enthusiastic s upport was very much appreciated. Finally, I would like to thank my mother who stood behind me unfailing through this arduous journey.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........9 LIST OF FIGURES................................................................................................................ .......11 ABSTRACT....................................................................................................................... ............12 CHAPTER 1 INTRODUCTION..................................................................................................................14 Study Purpose.................................................................................................................. .......14 Rationale for Study............................................................................................................ .....15 Study Aims..................................................................................................................... ........18 Significance of the Study...................................................................................................... ..21 2 MATERIALS AND METHODS...........................................................................................23 Background..................................................................................................................... ........23 Animals........................................................................................................................ ...........23 Animal Housing Conditions...................................................................................................24 Study Group Assignment........................................................................................................25 Pharmacological Agents.........................................................................................................25 Study Drugs.................................................................................................................... .25 Prednisolone Succinate............................................................................................26 Teriparatide..............................................................................................................26 Flurochrome markers.......................................................................................................27 Demeclocycline........................................................................................................27 Calcein......................................................................................................................27 Anesthesia and Euthanasia.....................................................................................................28 Bone Harvesting................................................................................................................ .....28 Study Measures................................................................................................................. ......29 Anthropomorphic Measures............................................................................................29 Histomorp homet ry...........................................................................................................29 Static bone measurements........................................................................................30 Dynamic bone measurements...................................................................................31 MicroCT........................................................................................................................ ..32 Femur.......................................................................................................................32 Vertebrae..................................................................................................................33 Statistical Analysis........................................................................................................... .......34

PAGE 7

7 3 LITERATURE REVIEW.......................................................................................................36 Bone Biology................................................................................................................... .......36 Structure of Bone.............................................................................................................36 Bone Cells..................................................................................................................... ..38 Osteoclasts................................................................................................................38 Osteoblasts...............................................................................................................39 Osteocytes................................................................................................................39 Bone Lining Cells.....................................................................................................40 Bone Remodeling............................................................................................................40 Remodeling Balance The RANKL/OPG/RANK Axis.................................................43 Effects of Glucocorticoid Drugs and Teriparatide on Bone...................................................44 Systemic Effects of Glucocorticoid Drugs......................................................................44 Direct Effects of Glucocorticoids....................................................................................45 Indirect Effects of Glucocorticoid Drugs........................................................................46 Decreased intestinal absorption of calcium..............................................................46 Increased renal elimination of calcium....................................................................47 Antagonistic action on gonadal functions................................................................47 Increased sensitivity to PTH....................................................................................48 Bone loss in response to glucocorticoid treatment..........................................................48 Parathyroid Hormone............................................................................................................ ..49 Dual Nature of PTH: Continuous versus Intermittent Administration...........................49 Parathyroid Hormone (PTH 1-84)...................................................................................49 Teriparatide (PTH 1-34)..................................................................................................51 Mechanisms of Action.....................................................................................................52 Studies of Glucocorticoid-Induced Bone Loss in Mice..........................................................53 Validity of the Mouse as a Model of Glucocorticoid-Induced Bone Loss......................53 Glucocorticoid-Induced Bone Loss in Mice....................................................................53 Teriparatide Treatment in Mice.......................................................................................54 Further Considerations......................................................................................................... ...55 4 RESULTS........................................................................................................................ .......56 Measurement Design............................................................................................................. .56 Anthropomorphic Measures...................................................................................................57 Bone Measures.................................................................................................................. ......58 Measurements of the Lumbar Vertebrae................................................................................58 Bone Volume/Total Volume...........................................................................................59 Trabecular Number..........................................................................................................61 Trabecular Width/Trabecular Thickness.........................................................................63 Trabecular Separation......................................................................................................66 Measurements of the Distal Femur.........................................................................................67 Bone Volume...................................................................................................................68 Trabecular Number..........................................................................................................71 Trabecular Thickness/Trabecular Width.........................................................................72

PAGE 8

8 Trabecular Separation......................................................................................................75 Osteoblast Surface and Osteoclast Surface.....................................................................76 Dynamic Measures of Bone Formation in the Distal Femur..................................................79 Mineralizing Surface.......................................................................................................79 Mineral Apposition Rate.................................................................................................81 Bone Formation Rate/Bone Surface (BFR/BS)...............................................................82 Mid-Shaft Cortical Bone Da ta and Significant Changes........................................................84 5 DISCUSSION..................................................................................................................... ....86 Glucocorticoid Drugs Suppressed Bone Form ation But Did Not Affect Bone Volume........86 Anabolic Effects of PTH Prevented the Inhibitory Changes Associated with Glucocorticoid Drugs..........................................................................................................89 PTH Increases Bone Mass Quickly........................................................................................91 Residual Effects of Glucocorticoid Drugs are Apparent During Natural Recovery..............92 PTH Treatment After Glucocor ticoid Use Was More Effec tive than Natural Recovery.......93 Age-Related Effects on Bone Mass........................................................................................94 Prophylactic Value of Concurrent Treatment with Glucocorticoid Drugs and PTH..............96 Site Specificity of GC and PTH Treatment............................................................................97 Conclusions.................................................................................................................... .........97 Clinical Applications.......................................................................................................... ....98 Study Limitations.............................................................................................................. ......99 Future Directions.............................................................................................................. ....100 APPENDIX A SUMMARY OF BONE MEASUREMENTS......................................................................101 B SUMMARY OF SELECTED STUDIES IN MICE.............................................................107 LIST OF REFERENCES.............................................................................................................111 BIOGRAPHICAL SKETCH.......................................................................................................121

PAGE 9

9 LIST OF TABLES Table page 3-1 Common Effects of Gl ucocorticoid Therapy.....................................................................45 4-1 Experimental Groups and Description of Treatments.......................................................56 4-2 Mean Animal Weights by Group.......................................................................................57 4-3 Mean Femur Lengths by Group.........................................................................................57 4-4 Summary of Histomorphomet ric Analysis of LV3 by Group...........................................59 4-5 Summary of MicroCT Analysis of LV2 by Group............................................................59 4-6 Significant Changes in Lumbar Vert ebra L3 Bone Volume by Group using Histomorp hometry .............................................................................................................60 4-7 Significant Changes in Lumbar Vertebra L2 Bone Volume by Group using MicroCT....61 4-8 Significant Changes in Lumbar Vert ebra L3 Trabecular Width by Group using Histomorp hometry .............................................................................................................64 4-9 Significant Changes in Lumbar Vertebra L2 Trabecular Thickness by Group using MicroCT........................................................................................................................ .....65 4-11 Summary of MicroCT Analysis of the Distal Femur by Group........................................68 4-12 Significant Changes in Distal Femur Bone Volume by Group using Histomorp hometry .............................................................................................................69 4-13 Significant Changes in Distal Femu r Bone Volume by Group using MicroCT................70 4-15 Significant Changes in Distal Femur Trabecular Number by Group using MicroCT.......72 4-16 Significant Changes in Distal Femur Trabecular Width by Group using Histomorp hometry .............................................................................................................73 4-17 Significant Changes in Distal Femur Tr abecular Thickness by Group using MicroCT....74 4-18 Significant Changes in Distal Fe mur Trabecular Separation by Group using Histomorp hometry .............................................................................................................75 4-19 Significant Changes in Distal Fe mur Trabecular Separation by Group Using MicroCT........................................................................................................................ .....76 4-20 Significant Changes in Distal Femur Osteoblast Surface and Osteoclast Surface by Group.......................................................................................................................... .......77

PAGE 10

10 4-21 Significant Changes in Distal Femur Mineralizing Surface by Group..............................80 4-22 Significant Changes in Distal Fe mur Mineral Apposition Rate by Group........................81 4-23 Significant Changes in Distal Femu r Bone Formation Rate by Group using Histomorp hometry .............................................................................................................83 4-24 Summary of Histomorphometric An alysis of Femur Mid-Shaft by Group.......................84 4-25 Summary of Significant Ch anges in Femur Mid-Shaft Cortical Bone Thickness by Group using MicroCT........................................................................................................84 A-1 Summary of Significant Changes in Lumbar Vertebrae L3 based on Histomorp hometry ...........................................................................................................101 A-2 Summary of Significant Changes in Lumbar Verteb rae L2 based on MicroCT.............102 A-3 Summary of Significant Ch anges in the Distal Femur based on Histomorphometry......103 A-4 Summary of Significant Changes in the Distal Femur based on MicroCT.....................104 A-5 Percent Changes in Osteoclast and Os teoblast Surfaces in the Distal Femur..................105 A-6 Percent Changes in Dynamic Bone Formation Parameters in the Distal Femur.............106 B-1 Studies Of Glucocorticoid -Induced Bone Loss In Mice..................................................108 B-2 Studies using Teriparatide in Mice..................................................................................110

PAGE 11

11 LIST OF FIGURES Figure page 2-1 Study Groups, Treatments and Timelines..........................................................................25 41 Lumbar Vertebra L3 Bone Volume/Total Volume by Group using Histomorp hometry .............................................................................................................60 4-2 Lumbar Vertebra L2 Bone Volume/T otal Volume by Group using MicroCT. ...............61 4-3 Lumbar Vertebra L3 Trabecular Nu mber by Group using Histomorphometry.................62 4-4 Lumbar Vertebra L2 Trabecula r Number by Group using MicroCT................................62 4-5 Lumbar Vertebra L3 Trabecular Wi dth by Group using Histomorphometry. .................64 4-6 Lumbar Vertebra L2 Trabecula r Thickness by Group using MicroCT.............................65 4-7 Lumbar Vertebra L3 Trabecular Sepa ration by Group using Histomorphometry. ..........66 4-8 Lumbar Vertebra L2 Trabecula r Separation by Group using MicroCT............................67 4-9 Distal Femur Bone Volume/Total Vo lume by Group using Histomorphometry. ...........69 4-10 Distal Femur Bone Volume/Total Volume by Group using MicroCT..............................70 4-11 Distal Femur Trabecular Number by Group using Histomorphometry.............................71 4-12 Distal Femur Trabecular Nu mber by Group using MicroCT............................................72 4-13 Distal Femur Trabecular Thickness by Group using Histomorphometry. .......................73 4-14 Distal Femur Trabecular Thic kness by Group using MicroCT. ......................................74 4-15 Distal Femur Trabecular Separati on by Group using Histomorphometry. ......................75 4-16 Distal Femur Trabecular Separa tion by Group using MicroCT. .....................................76 4-17 Distal Femur Osteoblast Surface by Group using Histomorphometry. ...........................78 4-19 Distal Femur Mineralizing Surf ace by Group using Histomorphometry..........................80 4-20 Distal Femur Mineral Apposition Ra te by Group using Histomorphometry....................81 4-21 Distal Femur Bone Formati on Rate/Bone Surface by Group using Histomorp hometry .............................................................................................................83 4-22 Mid-Shaft Femur Cortical Th ickness by Group using MicroCT.......................................85

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 Doctor of Philosophy SKELETAL EFFECTS OF TER IPARATIDE IN GLUCOCORTICOID-TREATED MICE By Kathleen S. Howe August 2007 Chair: Randy W. Braith Major : Health and Human Performance Synthetic analogs of glucoc orticoid (GC) drugs are wi dely used in treating many inflammatory diseases and conditi ons and are also used to suppr ess the immune system in solid organ transplant recipients. Ho wever GCs have serious side e ffects including osteoporosis and bone fractures. We conducted a randomized, pr ospective investigation of the effects of teriparatide in treating GC-induced osteopenia in mice and examined the extent and character of bone recovery in the distal femur and lumbar spine after exposure to GC when there was no subsequent treatment with teriparatide; simultaneous GC and teriparatide administrati on over the entire cour se of treatment; delayed administration of teriparatide Seven month old male Swiss Webster mice receive d prednisolone (2.1 mg /kg/d), teriparatide (40ug/kg/d), or vehicle to determine changes in bone structure and turnover after a 4or 8-week (6 d/wk) treatment regimen. We injected fluroc hrome markers (declomycin and calcein) before sacrifice and harvested femurs and lumbar vert ebrae to assess bone re sponse. Bone samples were analyzed using histomorphometry and mi croCT and both techniques showed the same trends.

PAGE 13

13 We found GCs suppressed bone turnover but not necessarily bone volume and that teriparatide, a bone anabolic agent, effectively increased bone turnove r and inhibited bone changes resulting from GC exposure. The effects of teriparatide were rapid and relative changes were greater in the distal femur than in the lumb ar spine. Four and ei ght weeks of teriparatide significantly increased both the osteoclast surf ace (Oc.S) and the osteoblast surface (Ob.S), resulting in significant increases in mineralizin g surface and mineral apposition rate. Increased Ob.S and Oc.S indicated the in creased turnover seen with teri paratide favored bone formation. We also detected a residual effect of GC on bone evidenced by lack of increased bone formation despite increased osteoblastic activity after GC treatment was discontinued. The underlying goal of our study was to demonstrate the efficacy of us ing PTH to prevent the adverse effects of GCs on bone in mice, as a prelude to studies in humans.

PAGE 14

14 CHAPTER 1 INTRODUCTION Glucocorticoid-induced bone loss is the leading cause of seconda ry osteoporosis (1). The negative effects of glucocortico id (GC) drugs on the skeletal system are well established but there is currently no consensus on the best way to prevent and/or treat the associated bone loss. We hypothesized that teriparatide, currently the only anabolic ag ent approved for the treatment of established osteoporosis, would be capable of reversing and/or prev enting glucocorticoidinduced bone loss. There are a number of patie nt populations that coul d benefit from such treatment. Chronic lung disease, rheumatic dise ases, and gastrointestinal diseases often involve prophylactic GC use. Patients awaiting solid orga n transplants would also benefit, since some transplant centers consider antecedent osteoporo sis a contraindication for transplant surgery because immunosuppressant regimens including GC cause rapid bone loss after transplantation. There have been studies examining the effects of GC use in mice (2-5) but no studies have examined the combined use of GC and teriparatide in an animal model. Furthermore, there has only been one study documenting the use of teripa ratide in humans exposed to long-term GC therapy (6). The limited use of teriparatide in humans treat ed with GC prompted us to select a mouse model for this study. We used the Swiss Webster strain of mice because they have significant levels of cancellous bone in the femur and spin e (7), have previously shown loss of bone in response to GC treatment (4,5), and an anabolic response to te riparatide (8). A mouse model allowed us to simulate chronic GC use. Study Purpose The purpose of this study was to determine the extent/character of bone recovery following exposure to GC drugs when there was

PAGE 15

15 No subsequent treatment with teriparatide Simultaneous GC and teriparatide administrati on over the entire cour se of treatment Delayed administration of teriparatide We used microCT and histomorphometric techniques to evaluate bone mass, bone resorption, bone formation, and microarchitect ural endpoints such as trabecular number, thickness, spacing, and connectivity density to determine whet her teriparatide improved bone mass and architecture following exposure to GC. This preliminary study, designed to assess the effects of teriparatide on gluc ocorticoid-induced osteopenia* in mice, was the first step in a research sequence that will help define new treatment options to improve the quality of life for clinical populations exposed to long-term GC therapy and thos e facing transplant surgeries. Rationale for Study Synthetic analogs of GC are a widely used class of drugs th at have proven effective in many inflammatory diseases and conditions, in cluding asthma, Chronic Obstructive Pulmonary Disease, rheumatoid arthritis, Cr ohns disease and lung diseases such as cystic fibrosis. GC analogs are also a key anti-rejection drug followi ng solid organ transplant ation (9,10). However, GC drugs pose serious side effects to the patient. Osteoporosis, with resu lting bone fractures, is the most incapacitating sequelae of GC therapy. B one is a dynamic, living tissue in which there is a normal balance of bone formation and bone resorption. This balance of bone loss and gain helps maintain a healthy skeletal structure capab le of withstanding normal loads and stresses. GC drugs disrupt the normal homeostasis of bone and rapidly lead to loss of bone mass and increased fracture risk. GCs have a negative effect on both the hard outer layer of cortical bone and the cancellous bone found next to the marro w. Although GC drugs affect both types of bone, the most profound and rapid effect s are seen in cancellous bone. Osteoporosis is a term based on T-scores established fo r human populations. As such, the term osteopenia rather than osteoporosis is used to describe decreased bone mass in animals.

PAGE 16

16 The relationship between long-term GC use a nd osteoporosis is well established and is often referred to as Glucocorti coid-Induced Osteoporosis (GIO) in humans (11-14). Deleterious effects on the bones are found in dosages often me t or exceeded in the treatment of many conditions. Dosages as small as 7.5 mg/day can resu lt in a loss of spinal trabecular bone of 9.5% in 5 months (15) indicating even low doses can ca use significant loss of b one. Bone loss occurs most rapidly in the first 6-12 m onths of treatment and appears to be dose and duration dependent (15-20). Osteoporosis has been reported in 50% of patients exposed to long-term GC treatment and spinal fractures occur at a rate 4-5 tim es that found in patients not treated with glucocorticoids (12,15). Fracture rates among thos e taking the drugs for more than five years approach 30% (21). In solid organ transplant patients, significant loss of bone mass can be detected as early as three months after transplanta tion (22). Bone mineral density losses average 5-15% during the first year and 1-2% annually subsequently ( 16,23). The significant morbidity and mortality associated with GIO makes its potential prevention or reversibility an important issue. Currently there is no established method for preventing GIO. A variety of anti-resorptive treatments have been tried, including calci um supplementation, bi sphosphonate agents, estrogenic and androgenic hormones, and calcitoni n, but none of these ha s proven effective in reversing the low bone formation that accompan ies long-term GC use (6). We found that calcium/vitamin D supplementation and nasal calc itonin can slow bone loss but is unable to restore lost bone mass (24). Targ eted resistance exercise and bisphosphonates have been shown to prevent spinal bone loss in solid organ tr ansplant patients but l ong-term compliance is problematic (10,25,26). A recent st udy involving healthy postmenopausal women showed that a combination of a bisphosphonate and a high impact exercise program increased bone mass more

PAGE 17

17 effectively than bisphosphonate treatment alone (27). However, follow-up testing 15 months after cessation of the interventi on showed bone gains had not been maintained. If these gains cannot be maintained in otherw ise healthy populations, it is un likely patients taking GC drugs will be able to do so. Teriparatide proved more efficacious in preven ting and/or reversing GIO. Unlike the antiresorptive drugs such as bisphosphonates, which only slow bone loss, teriparatide has an anabolic effect on bone, which may contribute to increased bone microarch itecture and strength. Studies using teriparatide have shown increases in bone dens ity at the femur neck and particularly the lumbar spine (28-31) and indicate increases are greater with teriparatide than with bisphosphonates (32,33). Fract ure reduction has also been associated with the use of teriparatide (29,31,34). A study of postmenopaus al women with one to two preexisting nontraumatic vertebral fractures showed that the ve rtebral fracture risk following teriparatide treatment was reduced by two-thirds and the rela tive risk of non-vertebral fractures was reduced by one-half (29). To date, there has only been on e study examining the efficacy of teriparatide in GIO in humans (6). However, the results of this study were confounded by a simultaneous use of hormone replacement therapy, which acts as an anti-resorp tive drug on bone. In that study, the combination of estrogen and teriparatide resu lted in significant bone de nsity increases in the axial skeleton but it is unclear whether teriparatide alone can ove rcome the deleterious effects of GC treatment. A number of studies have also shown that teriparatide stimul ates bone formation, increases cancellous bone volume, architec ture, cortical width, and biomechanical properties of bone in both mice (35) and humans (29). Teriparatide ha s been shown to increase cancellous bone and bone mineral density particularly in the axial sk eleton (8,35). Rodents are frequently used as a

PAGE 18

18 model for osteoporosis research because they ex hibit bone mass changes similar to humans when exposed to many osteoporosis-inducing stimuli (5). While the ovariectomized rat is the most commonly used animal model for postmenopausal osteoporosis, the mouse may be a better model for glucocorticoid-induced osteopenia (3,5 ) because researchers have found inconsistent responses to GC exposure suggesting rats may be re sistant to the effects of GC exposure (36-38). Studies in mice which have achieved skeletal matu rity (4,5) suggest the efficacy of the Swiss Webster strain of mouse in a glucocorticoid -induced model of bone loss (2,5,7). Bone loss patterns in mice exposed to GCs approximate hu man responses and the response to teriparatide in studies suggests the process is similar in both humans and mice, leading us to choose this animal for the study (3,5,39). There is also a growing recognition that along with bone mineral density (BMD), bone architecture should be examined to determine th e true efficacy of a treat ment (40,41). This study was designed to use both histomorphometry and microCT techniques to determine bone responses to GC treatment and whether there wa s any natural recovery following withdrawal of that treatment. We also, for the first time, de termined differing bone response to simultaneous treatment with glucocorticoid and teriparatide (p revention) versus subsequent treatment with teriparatide after bone loss ha s occurred (reversal). Study Aims Research Aim 1. To measure the changes in bo ne architecture and bone metabolism resulting from teriparatide therapy following, or in conjunction with, GC administration in a skeletally mature animal model. Hypothesis 1. In a skeletally mature mouse, teri paratide will reverse bone loss caused by 4 or 8 weeks of prednisolone treatment.

PAGE 19

19 Rationale: Mice experience dose-dependent loss of bone in response to GC treatment over a threshold level of 1.4 mg/kg for as little as 27 da ys although evidence of increased osteoclast numbers appears as ea rly at as 10 days of treatment (4,5). This study used a dose of 2.1 mg/kg body weight, consistent w ith other studies (4,5). Previous studies using mice have shown that chronic GC suppressed bone form ation leading to bone loss in both axial and appendicular skeletal sites (2-5). Research ers have noted increase d bone resorption (4), osteocyte apoptosis (5), and hi stomorphometric changes consistent with bone loss (2,3) in response to GC treatment. Subs equent treatment with alendr onate resulted in increased osteoclast apoptosis and preven tion of osteoblast apoptosis (4). However, although alendronate slowed GC-induced bone loss, it could not prevent it (4). Research Aim 2. To determine whether there are benefits to treating mice with teriparatide as a prophylactic measure by compar ing results when GCs and teriparatide are administered together versus using teriparatide after glucoc orticoid-induced bone loss has already occurred. Hypothesis 2. Starting teriparatide therapy at the same time as GC treatment will result in less bone loss than starting teri paratide after glucocorticoid-in duced osteopenia has developed. Rationale: Numerous studies in humans and anim als have shown that GC treatment has negative effects on bones. GCs direc tly affect bone cells at least in part by increasing osteoclast differentiation and activation le vels (4,5), and increasing oste ocyte and osteoblast apoptosis (4,5,42,43). Teriparatides anabolic effects on bone have the potenti al to slow or reverse these effects. Patients using teripa ratide after taking GCs for at least one year showed marked increases in bone density at the sp ine although these results were less evident at the hip (6). We believed a reversal of bone loss would also be seen in GC-treated mice also treated with

PAGE 20

20 teriparatide. We believed trea ting animals with teriparatide after GC exposure would likely attenuate bone loss and allow re building of some microarchit ectural features. Increased resorption following GC use can cause loss of trab eculae and we did not expect teriparatide to reverse this effect, since teri paratide can only build on existing bone. However, we believed the use of teriparatide after GC exposure would revers e some of the damage and that we would see a greater treatment effect when teriparatide treatm ent was started at the outset of GC exposure. We believed beginning teriparatide treatment concurrently with GC would result in less overall bone loss since teriparatide would offset the negative effects of the GC drugs. Research Aim3. To determine the extent of una ssisted recovery from GC therapy compared with recovery using teriparatide. Hypothesis 3. A therapy regimen consisting of 4 w eeks of prednisolone use followed by 4 weeks of teriparatide treatmen t will result in increased bone mass and improved architecture greater than will be seen from natura l recovery after withdrawal of GC. Rationale: We expected GC drugs to decrease bone mass and alter bone architecture. If GC use was halted, we believed bone metabolic pro cesses would return to baseline levels and there would be a gradual restorati on of at least some of the lost bone. We expected the natural repair process to be less robust, however, than if teriparatide treatment ha d been initiated after GC use. Teriparatide is highl y anabolic and increases bone tu rnover and alters metabolism in favor of bone formation. Alt hough teriparatide would be unable to replace lost trabeculae we expected to see increased bone formati on on the surface of existing trabeculae. Research Aim 4. Determine the degree to which skel etal responsiveness to teriparatide would vary by site (femur versus lumbar vertebrae) based on differences in the prevalence of trabecular bone at these sites.

PAGE 21

21 Hypothesis 4. The deleterious effects of GCs and bene ficial effects of te riparatide will be seen first and most extensiv ely in the vertebrae. Rationale: Human studies indicate th at the positive effects of teriparatide following GC treatment appear first and to the greatest degree in the lumbar vert ebrae (6). In the only study in humans to date examining the effects of teriparati de in GIO, increases in bone density were first detected in the spine (6). Ch anges in the hip were detected after 12 months of treatment, although these increases in bone density were det ected after teriparatide treatment had ended (44). Animal studies have yielded mixed result s with some finding the greatest changes in the vertebrae (2,8) and others finding the greatest change in the femu r (45). These differences may reflect postural differences in the models, since there is less mechanical loading on the spine in a quadruped. Nevertheless, we believed the most profound changes would be seen in the lumbar vertebrae because this site has more cancellous bone. Significance of the Study This study lays the groundwork for future human st udies. It is the firs t step in a research sequence that will help define new treatment options to improve the quality of life for patients exposed to long-term GC therapy and those faci ng transplant surgeries. This study used microCT and histomorphometry to compare changes in bone quantit y and microarchitecture with teriparatide treatment in an an imal model of GC exposure. This study also allowed further examination of the specific charac ter of bone loss caused by GC. Every year nearly 342,000 people in the United States die from lung diseases, which are the 3rd leading cause of death. Many of these patients are given GC drugs to combat the inflammatory conditions of their diseases. Pati ents suffering from cystic fibrosis, emphysema, and asthma remain on GC treatment for years. A study commissioned by the American Lung Association estimates that in the United Stat es there are 8.6 million patients suffering from

PAGE 22

22 chronic bronchitis and 3.1 million with emphysema. That study also found that 7.7% of adults and 8.8% of children and adolescents under age 18 suffer from asthma (46). With better medical treatments, these patients are surviving longer an d it is imperative we find a more effective way to treat the adverse effects of GCs on bone. Additionally, many end stage lung failure patients will be evaluated for possible lung transplant procedures. Some transplant centers view established osteoporosis as a contraindication to lung transplant surgery since patients will need to take a cocktail of immunosuppressant drugs after transplantation and these drugs have deleterious effects on bone. This makes finding an effective treatment for GIO imperative. In the U.S. there have been over 360,000 solid organ transplants since 1988. In 2005, 23,506 solid organ transplants were performed and 90,620 pati ents remained on waiting lists (47). Solid organ transplantation surgery is increasing over time. Advanc es in medical science have significantly increased survival times for these patients and preventing or reversing osteoporosis is becoming a more important quality of life issue.

PAGE 23

23 CHAPTER 2 MATERIALS AND METHODS Background This study was designed as a randomized, pros pective investigation of the effects of teriparatide in treating GC-induced osteopenia in mice. There is strong evidence that glucocorticoids have a deleterious effect on bone. Teriparatide is the only FDA-approved anabolic bone treatment currently available, but it has not been routinely used to reverse GCinduced bone loss. This study uses an animal model to assess the e ffects of the synthetic glucocorticoid methylprednisolone succinate (prednisol one) and teriparatide at the tissue and cellular level. To evaluate th e efficacy of teriparatide to prev ent or reverse GIO, 70 mice were randomized among 7 treatment groups receiving a co mbination of prednisolone, teriparatide, or vehicle to determine changes in bone structure and turnover at th e end of a 4-week or 8-week treatment regimen. At the end of the treatm ent regimen, bone samples were collected to determine changes in bone structure and architect ure. This protocol was reviewed and approved by the Institutional Animal Care and Use Committ ee (IACUC) at the University of Florida. Animals The study cohort consisted of 70 male, 7-mont h old, retired breeder Swiss Webster mice (Harlan Sprague Dawley, Indianapolis, IN). Th e Swiss Webster, an outbred mouse strain, was selected because they are known to have high leve ls of trabecular bone (7). Swiss Webster mice achieve peak bone density and cease longitudina l growth in the long bones between 5 and 6 months of age (5) prompting our use of 7-month old mice. Animals had at least seven days to acclimatize to minimize the effects of stress durin g shipment. Male animals were used to avoid the confounding effects of changes in estrogen levels through the li fecycle of female animals.

PAGE 24

24 We used retired breeders because the costs of feeding and maintaining animals until they reached the age necessary for the study woul d have been cost prohibitive. Animal Housing Conditions Animals were housed in the Special PathogenFree (SPF) facility of the Animal Care Services (ACS) department at the University of Florida. Animals were housed one per cage in micro-isolator cages that provided animals with filtered air. Cages were kept in a specially designed rack (American Caging Equipment, A llentown, New Jersey) c ontaining spaces for 70 animals (10 rows with 7 cages per row). Animal cages were moved each week so that animals rotated to different positions within the rack. Animals were maintained under standard care conditions with 12 hours light/12 hours dark in a climate controlled room with an average temperature of 21 de grees centigrade and humidity of 40%. Animals were fed a standard rodent chow, Teklad Irradiated LM485 mouse/rat chow (Harlan, Indianapolis, IN). This food was irradi ated with Cobalt-60 to kill any bacteria or viruses present. The rodent chow had a minimu m of 19% crude protein and 5% crude fat and a maximum of 5% fiber. The food contained 0.98% calcium and 0.66% phosphorus. Food intake was not controlled, but food was weighed ever y three days to determine each animals consumption. Food debris on the cage floor was not measured and was assumed to be comparable between cages. Water was avai lable ad libitum and was supplied through an automated watering system with the water purifie d by reverse osmosis. Animals were also provided with a supplemental water bottle. A ll procedures on animals were conducted under a hood to avoid exposing animals to contaminants. Animals were weighed using a digital balance at the end of their acclim ation period, weekly, and prior to sacr ifice. Body weights were used to monitor animal health.

PAGE 25

25 Study Group Assignment Study groups, treatments, and time lines are shown in Figure 2-1. START Weeks 1-4 Weeks 5-8 GROUP || (Baseline Sacrifice) BSL CNTL (n=10) | VEH | VEH ________ | 8 week Vehicle Cntl (n=10) | GC | SACRIFICE_______ | 4 week GC Cntl (n=10) | GC | GC | 8week GC Cntl (n=10) | GC | VEH | 4 week GC/Natural Recovery (n=10) | GC + PTH | GC + PTH | 8 week GC + PTH(n=10) | GC | GC +PTH | 4 week GC/ 4week GC+ PTH (n=10) Figure 2-1. Study Groups, Tr eatments and Timelines. Animals were the same age (7 months) at arrival and were block randomized to groups based on their arrival date at the ACS facility and their body we ight. Each week for seven consecutive weeks, 10 animals arrived and were distributed among treat ment groups with 1-2 mice/shipment/group. Pharmacological Agents Study Drugs In this study, prednisolone wa s used to model glucocorticoid treatment. After 4 weeks of exposure to prednisolone, the dr ug was discontinued in one group of animals to assess natural recovery of the bones. Other animals continued with glucocorticoid treatment alone or were simultaneously treated with teriparatide to see if that drug could prevent or reverse the effects of GC on the bones. Animals receiving teriparatid e received subcutane ous injection of 40ug/kg body weight/day, a dose commonly used in stud ies using mice (8,35). All study drugs and vehicles were formulated so that each animal received a subcutaneous injection volume of approximately 0.1 ml/injection. Animals were re strained by hand and each animal received two

PAGE 26

26 injections per day. Study drugs or vehicle was administered sequentially at approximately the same time of day throughout the study. All stud y drugs were prepared under sterile conditions. Prednisolone Succinate Prednisolone (Webster Veterinary Supply, St erling MA) or vehicle (sterile saline) was administered at a dose of 2.1 mg/kg/day, 6 days/w eek. The drug was purchased in liquid form at a concentration of 20 mg/ml and diluted with steril e saline. Prednisolone was prepared fresh 1-2 times per week under sterile conditions in a bioc hemistry lab in the College of Health and Human Performance. Teriparatide Teriparatide (Bachem, Torrence, CA) is a recomb inant PTH that consists of the same first 34 amino acids found in endogenous PTH. The an abolic action of the drug has been found to reside in this fragment (48). This drug is currently FDA-approved for use in humans to treat severe osteoporosis. Teriparatide was purchased in powder form and dissolved in an acidified, 2% heatinactivated mouse serum stock solution (vehicle) using a formulation used in previous rodent studies (49). Specifically, heat -inactivated mouse serum (obtained from adult, male Swiss Webster mice) was used to make a stock serum so lution that was used as vehicle and to dissolve teriparatide. The serum stoc k solution was prepared by mi xing 0.1 ml 0.001N HCL and 97.9 ml sterile saline. The sterile sa line and HCL was filtered using a 0.2 micron millipore filter and 2 ml of heat-inactivated serum was then added. The serum stock was divided into 1.5 ml aliquots and stored at -20 degrees C until needed to dilute the PTH stock solution or for use as vehicle. To prepare the PTH stock, 1 mg of teriparati de was diluted in 1 ml of the serum stock solution. The PTH stock solution was then divided into 10 l and 20 l aliquots and stored at 800 C until needed. Storing the PTH stock in sm all quantities ensured no aliquot was thawed

PAGE 27

27 more than twice. Dissolved teriparatide, was administer ed at a dose of 40ug/kg/day, 6 days/week subcutaneously in a volume of approximately 0.1 ml/mouse depending on body weight. Flurochrome markers Flurochrome markers (demeclocycline and cal cein) were injected to label actively mineralizing bone. Animals were injected subcutan eously on a pre-determined schedule prior to sacrifice. These flurochromes bind to calcium and are incorporated in to newly forming bone. They provided a means of determining the amou nt of bone mineralized between flurochrome treatments. This is a technique commonly used in histomorphometric analysis of bone formation. Demeclocycline Animals were injected with demeclocycline (Sigma, St Louis, MO) at a dosage of 15 mg/kg SC 11 and 10 days prior to sacrifice. This drug produces a dull orange fluorescent band on bone that can be seen under ul traviolet light when it binds to calcium in newly formed bone. Demeclocycline was purchased in powder form and dissolved in sterile saline. The mixture was stirred for at least two hours to in sure the demeclocycline completely dissolved. Demeclocycline was prepared fresh the morning of the 11th day prior to sacrifice and used for the -11 and -10 day injections. Any remaining volume was then discarded. Calcein Animals were injected with cal cein (Sigma, St Louis, MO) SC at a dosage of 15 mg/kg at 4 and 3 days prior to sacrifice. This drug binds to calcium in newly form ing bone and appears as a bright green band that can be s een under ultraviolet light. Ca lcein was purchased in powder form and prepared for injection by dissolvi ng it in sterile saline buffered with sodium bicarbonate. Calcein was prepared fresh the morn ing of the 4th day prior to sacrifice and used for the 4and 3-day injections. Re maining calcein was then discarded.

PAGE 28

28 Anesthesia and Euthanasia Animals were anesthetized using inhaled is oflurane (2-3.5%) with oxygen as the carrier gas using an anesthesia cart with a charcoal filter scavenger attached. Animals were placed one at a time into the anesthesia chamber. The isoflurane gas was started and the animals were observed until unconscious. They were then re moved from the chamber and deep anesthesia confirmed by a lack of motor responses to a pinc h of the foot. The animals were euthanized by exsanguination from the aorta, foll owed by cervical dislocation. Bone Harvesting Femurs and lumbar vertebrae were harves ted to assess (via histomorphometry and microCT) the bone response to treatment in th e appendicular and axial skeleton, respectively. Both femurs and vertebrae (13th thoracic through 5th lumbar, T13-L5) were excised from each animal. The femur was disarticulated from the acetabulum and the tibia using a scalpel. A scalpel was also used to shave off the cranial surface of the distal femur to expose the growth plate and metaphysis. The bone was then cut at about the mid-point usi ng a hand-held saw or bone shears. The lumbar vertebrae were harvested through an incision on the ventral si de of the animal. Internal organs were removed and the ventral por tion of the vertebral area gently scraped with a scalpel to allow visualization of the vertebrae an d intervertebral disks. The lumbar vertebrae were identified by first locati ng the floating ribs (T11 T13) A cut was made through the intervertebral disk cranial to the last thoracic ve rtebrae (T13). The area of the vertebral column to be removed was identified by counting interverte bral disks, which were visible as white bands on the ventral aspect of the spinal colum n. A second incision was made through the intervertebral disk caudal to the fifth lumbar ve rtebra. This allowed T13 and lumbar vertebrae L1-L5 to be removed as one section. Removal of T13 with the lumbar ve rtebrae made it easier

PAGE 29

29 to identify the cranial and cauda l ends of the lumbar vertebrae and, therefore, to identify individual vertebrae. The femurs and lumbar vertebrae were stor ed in 20-ml glass sc intillation vials in phosphate-buffered formalin for 24 hours. Af ter 24 hours, the formalin was poured off and replaced with 70% alcohol. The bones were then kept at 4 degr ees C until microCT and histomorphometric assessment. Study Measures Anthropomorphic Measures Animal weights were obtained using a digi tal scale (Ohaus Scout Pro, Pine Brook, New Jersey). Animals were placed in a weighing bucket to minimize movement and improve weighing accuracy. Animals were weighed afte r their acclimation peri od, weekly during the study, and prior to sacrifice. Animal weights were used to determine individual drug dosages and to monitor the health of the animals. The left femur from each animal was measured using an electronic di gital caliper (Little Machine Shop, Pasadena, CA) to confirm lack of longitudinal growth of the femur over time among groups. Each bone was measured twice and the average of these measurements was used. The electronic caliper was zeroe d between each measurement. Histomorphometry Bone specimen preparation for histomorphometric analysis was carried out at the Wronski Lab, Department of Physiological Sciences, Univ ersity of Florida usi ng established protocols described elsewhere (50,51). In brief, the right femur and lumbar vertebrae L3 were dehydrated in increasing concentrations of ethanol over a 1week period and cleared in xylene for 24 hours. The samples were then embedded undecalcified in modified methylmethacrylate to facilitate sectioning. The embedding pr ocess involved treating the bon es in a series of four

PAGE 30

30 methylmethacrylate solutions (w ith increasing amounts of a cat alyst) that progressively infiltrated the bone over a period of 9 days. Th e bones were then placed uncapped in a vacuum dessicator for 6-8 hours. Subse quently, the bones were positioned in the center of the vial to optimize the sectioning process and placed in a wa ter bath at 42 C over night. The heat caused the methylmethacrylate to polymerize and hard en. Once the methylmethacrylate hardened, the glass vial was broken and removed leaving a plasticized block containing the bone. The embedded bones were sectione d longitudinally at 4and 8 m thickness using Leica/Jung 2050 or 2165 microtomes. Six non-consecu tive 4 m sections and 6 non-consecutive 8 m sections were cut and mounted on gela tinized glass slides. The two best 4 m and two best 8m sections were selected for analysis. The 4 m sections were stained for assessment of static (structural and cell) measurements while the 8 m sections were coverslipped unstained for evaluation of dynamic measurements. Static bone measurements The 4 m thick sections were stained according to the Von Kossa method with a tetrachrome counterstain (Polysciences Inc ., Warrington, PA) (50). This stain causes mineralized bone to appear black and bone cells and osteoid to stain blue. Static structural and cellular e ndpoints were measured in two 4 m stained sections using the Trabecular Analysis System (TAS)/Osteomeas ure System (Osteometrics Inc., Atlanta, GA) or the Bioquant Elite Bone Morphometry System (R&M Biometrics, Nashville, TN). Endpoints measured or calculated included Cancellous bone volume/tota l volume (BV/TV, %) (percen tage of total marrow area occupied by cancellous bone) Trabecular width (Tb.Wi, m) (1.99 x B ar/2/ b Pm) Trabecular number (Tb.N, #/mm) (BV/TV)/Tb.Th)

PAGE 31

31 Trabecular separation (Tb.Sp,m) ((1/Tb.N) Tb.Th) Osteoblast surface/bone surface (O b.S/BS, %) (percent of bone surface lined by osteoblasts) Osteoclast surface/bone surface (Oc.S/BS, %) (p ercent of bone surface lined by osteoclasts) For evaluation of structural endpoints using TAS, the bone section was magnified 2x and a video capture system was used to take an image of the bone. The region of interest (ROI) is 1.5 mm2, beginning 0.5mm proximal to the growth plate and extending back toward the diaphysis. The ROI was also 0.25 mm from the cortical bone on either side of the femur. TAS software allowed the user to modify the video image to match the bone section and the software then calculated the amount of bone presen t within a region of interest. This data was then used to calculate Tb.N, Tb.Th, and Tb.Sp as defined above. Ob.S and Oc.S were measured using the Bioqua nt Elite Bone Morphometry System. This system allowed us to manually trace the total peri meter of cancellous bone as well as the portions of cancellous bone surface covered by osteoblasts and osteoclasts to calc ulate the proportion of the total cancellous bone covered by these cells. Dynamic bone measurements Dynamic bone analysis was accomplished using flurochrome-based data collected from unstained 8 m femur sections using the Osteomeas ure System. Two sections from each animal were used and the results averaged. Fl urochrome data was used to determine Mineralizing surface (MS/BS, %) (percentage of cancellous bone su rface with a double flurochrome label; MS/BS is a dynamic index of bone formation). Mineral apposition rate (MAR, m/day) (dis tance between the two flurochrome labels divided by the number of days between label ad ministration; MAR is an index of osteoblast activity). Bone formation rate/bone surface (BFR/BS, um3/um2/day) (MS x MAR; volume of new bone formed per unit of total bone surface per unit time)

PAGE 32

32 The slides were magnified at 200X on the microscope and displayed at 250X on the computer monitor. The area of interest was de fined as the area beginn ing approximately 0.5 mm proximal to the end of the growth plate and consists of a series of fields that, combined, equal a cancellous bone area approximately 1.5 mm X 1.5 mm that is a bout 375 m from the cortex. The cancellous bone (with and without flurochrome labels) was outlined using a digitizing tablet. Then, the inner and outer flurochrome labels we re outlined where double labeling exists and the distance between these two lines was measured at 4 approximately equidistant points. The software then calculated the MS, MAR, and BFR/BS as defined above. MicroCT MicroCT was used for nondestructive th ree-dimensional evaluation of bone microarchitecture. MicroCT identified subtle changes in three-dimensional bone architecture that cannot be detected by histomorphometry. The bones we re scanned using a Scano microCT40 scanner (Scanco Medical AG, Basserdor f, Switzerland). Cancellous bone in the LV and femoral metaphysis was evaluated. Femur Prior to placement in the microCT, the fe murs were first cleaned of non-skeletal connective tissue and muscle and placed between tw o thin Styrofoam pads. The Styrofoam pads help kept the samples from m oving during the test. The bones and padding were placed in a specially designed tube that was 12.3 mm in diamet er. Three femurs were loaded into the tube and scanned sequentially. The t ube was then filled with 70% ET OH and covered with parafilm. The samples were scanned at medium resolution at a voxel size of 12.3 x 12.3 x 12.3 m. Scanning took approximately one ho ur per bone. Reconstruction of the bones following the scan took an additional hour per bone. The volume of interest in the distal fe mur consisted of 1.8 mm2 starting at the growth plat e and moving toward the diap hysis. Of the 120 180 slices

PAGE 33

33 scanned, 150 were analyzed. Twenty slices (app roximately 0.25 mm) of co rtical bone were also analyzed in the femoral midshaft. The VOI for cortical bone began at th e midpoint of the femur and included 55 slices toward the proximal femur. Of the 55 slices scanned, 20 were analyzed. Direct cancellous bone measurem ents in the femur included: total tissue volume (combined volume of can cellous bone and bone marrow in the volume of interest (VOI) Cancellous bone volume (volume in the VOI occupied by cancellous bone) Trabecular thickness Trabecular number Trabecular separation Cortical thickness was measured in a sample from the mid-shaft Once the bone was scanned and reconstructe d, drawing tools provided with the software were used to outline the area of interest (AOI). Every tenth slice was contoured by hand and the software extrapolated the AOI to the remaining slices. A visual inspection of each slice was done to ensure no cortical bone was included in the cancellous ROI. The ROI for the distal femur scanned consisted of the cancellous bone pr oximal to the growth plate extending to about 1.8 mm toward the diaphysis The cortical bone analyzed began at approximately the center of the diaphysis to a point 20 uCT slices (250 m) toward the distal metaphysis. Vertebrae The second lumbar vertebrae (L2) was analyzed using uCT. First, L2 was separated from the rest of the vertebrae a nd non-skeletal connective and muscle tissue was removed. The vertebrae were scanned eight at a time. The sp inal canal of each vertebra was threaded onto a slender wooden holder and a small piece of Styr ofoam was placed at the bottom of the wooden holder, between each vertebrae, and at the top of the holder. This held the vertebrae upright and

PAGE 34

34 helped controlled movement during the scan. A scout scan was run to identify the area for analysis and then the scan proceeded automati cally. The samples were scanned at medium resolution at a voxel size of 12.3 x 12.3 x 12.3 m. The scan took approximately one hour per bone and required an additional hour per bone recons truction time prior to analysis. As with the femoral metaphysis, cancellous bone measur ements in the vertebrae included Total tissue volume Cancellous bone volume Trabecular thickness Trabecular number Trabecular separation. The AOI for the vertebra included all of th e secondary cancellous bone between the two growth plates. Once the bone was scanned a nd reconstructed, the software drawing tools provided were used to contour the AOI at every sixth slice. The software then extrapolated the AOI to the other slices. All slices were reviewed to ensure no cortical bone was included in the AOI. Statistical Analysis A power analysis was conducted to determine adequate sample size using data from a study evaluating the effects of tr eating Swiss Webster mice with the same prednisolone dosage proposed for this study (5). The present study wa s designed to achieve a power of at least 0.80. Power analysis using SAS version 4.0 (SAS Instit ute Inc., Cary NC) indicated eight animals per group would result in a 0.87 power. Each group in this study contained 10 animals to insure study power was not compromised if any animals ha d to be removed from the study prematurely. Data is presented in table format as mean sta ndard deviation (SD) and bar graphs with mean standard error for continuous variables, and as percent change for st atistically significant differences between groups. Statistical an alysis on data was c onducted using SPSS 10.0

PAGE 35

35 statistical software (SPSS Inc, Chicago, IL). Data was analyzed using the nonparametric Kruskal-Wallis test (52). When significant tr eatment differences were observed, between-group comparisons were performed using the Mann Whitney test of independent samples.

PAGE 36

36 CHAPTER 3 LITERATURE REVIEW This chapter is divided into two parts. Part A will present a ge neral overview of bone structure and metabolism, including a look at the various types of bone cells that contribute to the remodeling process. The four phases of bone remodeling: activation, resorption, reversal, and formation are also described. In the remodeli ng process, bone is first resorbed and then new bone is deposited. There is still much we do not understand about this pr ocess and how lifestyle, health, and pharmacologic interventions can influe nce the balance. The second part of this chapter describes the effects of two pharmacological agents, one catabolic (prednisolone) and one anabolic (teriparatide), on bone remodeling. Normal levels of remodeling are altered by both of these drugs, but in differe nt ways, and Part B of this ch apter will describe what we know about the influences of thes e drugs on the mouse model. Bone Biology Structure of Bone There are two basic types: cor tical and cancellous. Cortical b one is the hard outer layer of bone and is denser than cancellous bone, which is found closer to the bone marrow. By volume, cortical bone makes up about 80% of the adult hu man skeleton. The remaining 20% is the more changeable cancellous bone. In contrast to the hard and only slightly porous cortical bone, cancellous bone is a complex three-dimensional network of curved plates and rods in close association with bone marrow and is enclosed by cortical bone. Cancellous bone is made up of a lattice of large plates and rods collectively called trabeculae. Th e inner or endocortical side of cortical bone within th e medullary cavity (53). Cancellous bone is mainly found in bones of the axial skeleton, in flat and irregular bones, and in the ends of the long bones (53). This type of bone experiences deformation when loaded

PAGE 37

37 and is better able to bear loads without beco ming damaged (54). The lattice-like structure of cancellous bone means its surface-to-volume ratio is higher than that of cortical bone. Since remodeling takes place at the su rface of bones, the greater relativ e surface area of cancellous bone means remodeling takes place there at a rate te n times greater than in cortical bone (53,55). Thus, when there is an imbalance leading to mo re bone resorption than formation, the effects will be most apparent in cancellous bone such as that found in the vertebrae and the ends of long bones. Bones do not contain cortical and cancellous bone tissue uniformly. A typical long bone has 3 regions that vary in composition. The dia physis, the shaft of the bone, is comprised mainly of cortical bone. The ends of the long bones, known as the epiphyses, are to a large degree cancellous bone, as are the metaphyses, the conica l section of bone connecting the epiphysis and diaphysis. Cancellous bone is ma de up of plates of bone tissue a nd loss results in a gradual shift from plate-like to rod-like structures as the dominant elements (56). This contributes to increased fragility of the bone, since spaces with in the bone increase as the struts connecting one section to another disappear. Electron micr oscopy of an older persons bones shows wider spaces and fewer structural connections. This has important implications for bone strength since trabecular struts, once lost, ca nnot be replaced (56). Trabecu lar compromise occurs when osteoclasts erode a cavity too deep ly or when the osteoblasts are unable to lay down a sufficient amount of replacement bone (56). The major components of cortical and cancel lous bone are type 1 collagen, water, hydroxyapatite mineral, and small amounts of prot eoglycans and noncollagenous proteins. Type 1 collagen is a structural protein found mainly in bone and tendons. Hydroxyapatite, Ca10(PO4)6(OH)2, makes up virtually all of the minera l in bone and represents the major

PAGE 38

38 storehouse for the bodys calcium. Calcium is take n in or released base d on fluctuating plasma calcium levels and the presence of the major cal cium regulating hormones, PTH and calcitonin. Bone Cells Bone cells include osteoclasts, osteoblasts, osteocytes, and bone-lining cells. Each cell type is critical to bone rem odeling and these cells have complex mechanisms of communication that control their actions and interactions. Th ese cells generally operate in balance with each other, although aging, some disease conditions, and cer tain drugs can alter that balance. There is ample opportunity for the balance in bone metabo lism to shift toward more resorption than deposition since the osteoclast is able to resorb in one day an amount of bone that osteoblasts need several days to replace. Osteoclasts Osteoclasts are large, multinucleated cells a ssociated with bone resorption. Osteoclasts originate from hematopoietic stem cells in the bone marrow and travel via the circulatory (or perhaps the lymphatic) system. Mature osteocla sts are responsible for bone resorption, where bone is broken down and the calcium within libera ted. Osteoclasts adhere to bone by means of an actin ring that is anchored to the extracellular matrix by integrins. This forms a sealing zone that creates a microenvironment between the osteoc last and the surface of the bone that will be resorbed (57). When osteoclasts attach to bone the cell is polarized and generates a ruffled border. It is at this ruffled border that vesicles containing cathepsin K and membrane-bound H+ ATPase exist. Cathepsin K, an acidic collage nase, degrades the orga nic component in bone (type 1 collagen) while the H+ ATPase secretes hydrochloric acid into the sub-cellular space and dissolves hydroxyapatite (58). Osteoclasts are found in cavities on the bone su rface, which they themselves form, called resorption pits or Howships lacunae. Interestingly, alt hough associated with resorption,

PAGE 39

39 osteoclasts have no receptors for PTH, the main endogenous mediator of bone breakdown. Instead, osteoclasts have receptors for calcitoni n, a hormone that inhibits bone resorption (59). Osteoblasts Mature osteoblasts are bone-forming cells th at typically reside on the bone surface where they secrete unmineralized matri x, called osteoid, during the bone fo rmation process. They also participate in calcification of bone and regulate the movement of calcium and phosphate into and out of the bone. These cells are normally cuboi dal in shape. Osteobl asts themselves produce and secrete a number of substances important to bone metabolism including type 1 collagen, non-collagenous matrix proteins such as os teocalcin and osteonectin, growth factors, prostaglandins E1 and E2, Receptor Activator of Nuclear factor kappa B ligand (RANKL) and osteoprotegerin (OPG) and cytokines including interleukin (IL)-1, IL6, and 11, TNF, and TGF(60). Although osteoblasts are most noted fo r bone formation, they also help control bone resorption since they have receptors for PTH and secrete OPG and RANKL. In humans, osteoblasts secrete osteoid at the rate of about 1 micrometer per day. This means that either the lifespan of the osteoblast is quite long or that multiple generations of osteoblasts are involved in ref illing a given resorpti on pit since these pits can be quite deep (61,62). When the osteoblast has finished secr eting osteoid, it return s the preosteoblast pool, transforms into a bone-lining cell, gets bur ied as an osteocyte, or dies (53). Osteocytes Osteoblasts which become trapped in the osteoi d they secrete are called osteocytes. Each lacunae contains only a single osteocyte. These cells maintain contact with each other and with bone-lining cells via slender proc esses that reach through the canal iculi of the bone at the gap junctions. There are gap junctions between ad jacent bone-lining cells and between bone-lining cells and osteocytes. Osteocytes are thought to be involved in detecting microfractures and the

PAGE 40

40 cell signaling that begins the pr ocess of remodeling (53,61,62). Th ey may also be involved in storing mineral ions following a meal rich in cal cium and in transporting minerals from deeper skeletal reservoirs to the extracellular fluid compartment after resorption (53). Bone Lining Cells The final major bone cell type is the bone-lini ng cell. These are long, flat cells that cover quiescent (or resting) bone surfaces, where bone is neither being resorbed nor formed. Like osteocytes, bone lining cells orig inate from osteoblasts. They differ from osteocytes, however, in that they remain on the bone surface rather than being buried in the matrix. As bone formation ends, bone lining cells remain on the newly formed bone surface. They communicate with osteocytes and each other th rough gap junctions. The bone lin ing cells assist the osteocytes in moving mineral in and out of the bone and may also play a role in sensing mechanical strain on bone (63). Bone Remodeling Bone remodeling is the term used to descri be the processes of resorbing old bone and depositing new bone at the same site. Bone re modeling is an on-going h ousekeeping activity of healthy bone. Even in the absence of external stimuli there will be remodeling activity. This process of remodeling is most obvi ous in its accelerated form when there is a fracture and the body quickly moves to repair the damage. On a more subtle scale, however, the body is constantly replacing old bone and repairing micr ofractures, the damage caused by daily activity. Repair of this damage helps keep the bones st rong by preventing structural weaknesses from accumulating. Continual remodeling helps bone main tain strength and structural integrity, so long as there is a balance between resorption and deposition. In the aggregate, what determines whether more bone is being formed or resorbed is the relative amount of each activity. Hormones such as estrogen, growth hormone, insu lin, parathyroid hormone (PTH), testosterone

PAGE 41

41 and agents like fluoride and alumi num directly or indirectly affect the balance to varying degrees (53,61,62,64,65). Basic Multicellular Units (BMUs) orchestr ate bone turnover, re moving mechanically unneeded bone and repairing microdamage by laying down new bone. BMUs consist of osteoclasts and osteoblasts that congregate at a specific area of the bone where they have been drawn, possibly by the signaling activity of osteocyt es (66). In cortical bone the BMU tunnel in a cone-like pattern through the bone while in th e trabecular bone BMUs scallop the surface of the trabeculae to form a trench (39,43). The BMU remodels bone in four distinct phases: activation, resorption, reversal, and formation. When not involved in remodeling, bone is said to be quiescent. Bone remodeling begins when a quiescent skel etal surface is activated. The activation phase is characterized by a retr action of the bone lining cells at the activati on site and formation of new blood vessels that will bri ng osteoclasts to the resorption site. This exposes mineralized bone surface which may act as a chemoattractant for osteoclast precu rsor cells (53,61,62). During the activation phase, preosteoclasts fuse to form the characteristic multinucleated mature osteoclasts which will att ach themselves to the e xposed bone surface (43). The resorption phase begins when the multinucle ated osteoclasts begin to resorb bone at the remodeling site. The osteoclast, with its ruffled border, attaches to the area of the bone to be resorbed; here the osteoclast and the underlying bone form the microenvironment into which the osteoclast secretes acid s. When the osteoclast secretes aci ds into this microenvironment, the collagen matrix breaks down, forming concave pits called resorption pits or Howships lacunae. In humans, the resorption pits have an averag e erosion depth of 60 mi crometers in trabecular bone and about 100 micrometers in cortical bone. The osteoclasts can erode up to tens of

PAGE 42

42 micrometers per day (61). The whole process of resorption takes 1-3 weeks and culminates with the release of calcium and other compounds from the matrix of the dissolved bone. The end of resorption is marked by osteoc lasts migrating from the bone surface to nearby marrow spaces, where they hibernate or die (67). The third stage of remodeling, reversal, is ch aracterized by prepara tions to lay down new bone. Phagocytes smooth out ragged edges left by the osteoclasts and a thin layer of collagen and matrix, referred to as a cement line, is laid down. Osteoblasts are drawn to the area through as yet not understood mechanisms. Osteoblasts may be stimulated to mature by signals from compounds such as growth factors released from the bone itself when it breaks down. Some believe these signals occur when the calcium re leased by resorption activates calcium receptors on osteoblasts. This, theoretically, helps insure b one resorption does not get out of control, since the more resorption there is, the more osteoblasts would be stimulated to form new bone (68). Once at the remodeling site, osteob lasts adhere to the cement line wh ere they begin filling in the resorption cavity with osteoid (43). In the final phase of remodeling, known as formation, new osteoid is secreted by the osteoblasts. Under normal conditions, if there is sufficient calcium available, the new bone is mineralized. Fluroc hrome labeling is often used to measure this process. In humans, the entire sequence of resorption a nd formation at a given remodeling site takes place over a period of several months and it is estimated that the li fespan of a BMU is about 6-9 months (43,64,68). In healthy adults between 3 and 4 million BMUs are activated annually, with about 1 million operating at any given time (43). The remodeling cycle in animals follows the same sequence of events but the time required is significantly less. Th is accelerated response

PAGE 43

43 in animals makes them an attractive model to pred ict the effects of conditions and treatments in humans Remodeling Balance : The RANKL/OPG/RANK Axis The balance of bone remodeling is controll ed by hormones and paracrine influences originating from osteoblasts or stromal cells The discovery of RANKL and OPG was an important milestone in bone research since this helped explain a seeming paradox of bone remodeling. Namely, that many of the hormones, cytokines, and growth factors that regulate osteoclast activity have receptors on the osteoblast (69). Researchers had also noted that cell cultures of osteoclast precurso rs physically separated from os teoblasts did not develop into functional osteoclasts (65) and osteoclast apopt osis increased, indicati ng a close relationship between osteoblasts and the production and differe ntiation of osteoclasts. RANKL is produced by osteoblasts and binds to RANK receptors on os teoclasts and osteocla st precursors where it stimulates differentiation and greater activity of os teoclasts. OPG is also produced by osteoblasts but is a decoy r eceptor that competitively binds RANKL. The amount of bone resorption is modulated at le ast to some extent by the ra tio of RANKL to OPG. It is the ratio of RANKL to OPG, and not just the level of RANKL, that seems to govern whether bone remodeling favors formation or resorpti on so it is important to see what substances influence the ratio (70). OPG production is stimulated by 1,25 dihydroxy vitamin D3, BMP-2, TNF, IL-1 and -1 and estrogen (1,71). The resulti ng increase in OPG production removes additional amounts of RANKL, thus decreasing the amount of RANKL available to bind to RANK which results in reduced osteoclast differentiation and ac tivity. Many circumstances lead to increased RANKL production, such as glucocorti coid use, lack of es trogen, and it is often present in diseases like rheumatoid arthritis.

PAGE 44

44 Effects of Glucocorticoid Drug s and Teriparatide on Bone GCs inhibit the formation of osteoblasts and os teoclasts, increase apoptosis in osteoblasts, and interfere with normal bone remodeling. Wh ile the negative effects of GCs on bone have long been recognized, the mechanisms remain to be fully understood (72-75). Glucocorticoids bind to a cytoplasmic glucocorticoid receptor (GR) found on osteoblasts (76). The receptor is a ligand-operated transcription fact or. When not bound, the receptor is located in the cytoplasm as a protein complex. When activated, the complex dissociates and the receptor moves into the nucleus and binds to regulatory elements in th e promoter regions of cer tain anti-inflammatory genes (77-79). The GR also inactivates infla mmatory genes by binding to transcription factors activator protein-1 (AP-1) a nd nuclear factor kappaB (NF-KB) (77,78) With these transcription factors bound there is inhi bition of pro-inflammatory cytokines such as IL-1 IL-4, IL-5, and IL-8, and TNF(77). Genomic effects begin probably no sooner than 30 minutes after GC administration, and are initiated by binding of the steroid to cytosolic receptors. Nongenomic effects occur sooner, often within a few mi nutes, and are mediated by membrane-bound GC receptors (74). There appears to be general ag reement that GCs cause decreased bone formation; the case is not so clear for bone resorption (21,23). Systemic Effects of Glucocorticoid Drugs GCs exert a number of effects, both direct a nd indirect, that influence bone metabolism. (see Table 3-1). The direct effects of glucocortic oid drugs are those that effect the bone cells themselves and includes actions that effect osteoc lasts, osteoblasts and oste ocytes. Direct effects also include influences on the production of RANKL and OPG and influences on various bonerelated growth factors. Indire ct effects include influences on organ systems that influence calcium metabolism.

PAGE 45

45 Table 3-1. Common Effects of Glucocorticoid Therapy. Direct Effects Indirect Effects Increased Osteoclast formation Increased Urinary Calcium Excretion Increased Osteoblast apoptosis Increased Intestinal Calcium Absorption Increased Osteocyte apoptos is Decreased GH production Increased RANKL Hypogonadism Decreased Osteoblast #s and activity Impaired renal function Decreased OPG Secondary Hyperparathyroidism Decreased Type I Collagen Production Decreased skeletal growth factors (IGF-1, TGF) RANKL = Receptor activator of nuclear factor (NF)-kB ligand; OPG = Osteoprotegerin; IGF-1 = Insulin-like Growth Factor; TGF= Tumor Growth Factor; GH= Growth Hormone. Direct Effects of Glucocorticoids The effects of GCs on osteoblas ts are potent, causing pre-oste oblasts to differentiate to adipocytes and decreasing synthesi s of type I collagen by mature os teoblasts (80,81). GCs also decrease the ability of osteoblasts to adhere to the extracellular matrix and promotes matrix breakdown by stimulating the activity of interstitia l collagenase. Additionally, GCs amplify the response of osteoblasts to endogenous PTH by in creasing the number of PTH receptors on the cell (23). The osteoblast lifes pan decreases, leaving less time for synthesis of bone matrix and mineralization. Taken together, these effects result in a significant decrease in bone formation as evidenced by sharp reductions in circulating leve ls of osteocalcin even at low doses (~ 5mg/day in humans) of GC (76,82). GCs also inhibit a number of growth factors such as IGF-1, which increase the synthesis of t ype I collagen, and decrease colla genase 3 expression (80). Although there is general agreement that GC us e decreases bone forma tion, it is less clear whether GCs increase bone resorption (23,74,83). St udies of bone cells in vitro have variously shown stimulation and inhibition of osteoclasts in ce ll cultures (23,83) in response to GC, and decreased apoptosis of mature osteoclasts (75) Some histomorphometr ic studies have found increased resorption (84) in th e presence of GCs but serum and urine markers of bone resorption have shown inconsistent results (84).

PAGE 46

46 GC treatment increases osteoblast producti on of RANKL and col ony-stimulating factor (CSF)-1 (also known as macrophage-colony stim ulating factor or M-CSF) (81). The combination of M-CSF and RANKL stimulates oste oclastogenesis. At the same time RANKL is increasing, OPG levels decrease. GC drugs ha ve been shown to inhibit OPG mRNA by 70-90%, increase mRNA levels of RANKL and RANKL/ M-CSF-induced TRAP activity by over 50% (1,60,85). This has the potential to shift the bone remodeling balance in favor of resorption by increasing osteoclast formation. With GC use in humans, there appears to be an early increase in bone loss which moderates over time, creating a biphasic re sponse (76,80,86,87). One explanation for early increases in resorption that subside later is the influence of GCs on induction of IL-6 receptors in bone (87). Since IL-6 is a cyto kine important in osteoclast r ecruitment, any increase in the number of receptors in skeletal tissue c ould increase bone resorp tion by recruiting more osteoclasts. At the same time, GCs also inhi bit osteoblastogenesis. Declining numbers of osteoblasts eventually will pr oduce less aggregate RANKL, causing reduced osteoclastogenesis as well (88). That may explain observations th at the greatest bone losses from GC use are experienced early in treatment and the rate of loss decreases and le vels off over time. Indirect Effects of Glucocorticoid Drugs In addition to the direct effects of GC on bone metabolism there are also indirect effects that similarly result in bone loss over time. The indirect effects of GCs on bone involve a number of organ systems in the bod y and are summarized in Table 3-1. Decreased intestinal absorption of calcium GC use results in a decrease in calcium ab sorption in the intestin es (80). Although the mechanism is not entirely understood, GCs app ear to effect the duodenum by inhibiting active

PAGE 47

47 calcium transport, decreasing the production of calcium-binding proteins and possibly increasing the degradation of 1,25(OH)2 vitamin D at its binding site (21,74,76,86). Increased renal elimination of calcium Increased renal excretion of cal cium may be due to a reductio n in reabsorption of calcium in the distal tubule of the kidne y (74,76). In the presence of GCs, the kidney tubules handle sodium and calcium cations diffe rently (86). GC treatment increases activity of epithelial Na+ channels, passive sodium channels on the apical membrane of the dist al tubules and conducting duct cells and this increases the activity of Na+/ Ca2+ antiport pumps, resulting in increased calcium extrusion (86). Increased renal calciu m excretion coupled with decreased intestinal absorption may lead to secondary hyperparathyroi dism (23,74). It is unc lear how this affects overall bone remodeling, however. It was once believed that secondary hyperparathyroidism accounted for GC-mediated changes in bone, but re search now suggests the situation is much more complex and dynamic (72). Even in cases where secondary hyperpar athyroidism occurs, it does not explain the trabecular bone loss seen with GC use (80). In patients with secondary hyperparathyroidism, bone remodeling is increased ( 80) and the main effects are seen in cortical bone (81) instead of the decrease d remodeling that primarily affe cts cancellous bone as seen in GIO. Antagonistic action on gonadal functions Researchers have identified a direct GC -mediated effect on the production of gonadal steroids in men and women (74). It is believed GCs suppress the hypothalamic-pituitary-adrenal axis and inhibit gonadotropin s ecretion (84). GC treatment has been shown to decrease circulating levels of testoster one in men by about 50% (75). Similar effects on estrogen are believed to occur in women (18,75). Both estr ogens and androgens suppress bone resorption by

PAGE 48

48 inhibiting osteoblastic release of local stimula ting factors that cause formation of increased numbers of osteoclasts (17,75). Increased sensitivity to PTH In vitro studies of isolated bone cells have shown that GCs modulated PTH sensitivity of both osteoblasts and osteocytes such that lower levels of PTH still elicited measurable biochemical changes (18). This may be accomplished by GC-mediated upregulation of osteoblast PTH receptors (74) or increased affin ity of the receptor for PTH (23,81). This could explain why changes in bone are seen even when PTH levels remain in the normal range. Bone loss in response to glucocorticoid treatment Studies have shown that with GC treatment, there is a loss in trabecular connectivity making this population more susceptible to fracture (23,76,81). This change in bone microarchitecture cannot be detected with dens itometry, the most common clinical means of testing for bone loss. Some have suggested that the GC-mediated lo ss of osteoblasts and osteocytes compromise bone streng th independent of bone loss (72). According to this theory, the integrity of bone re lies on the network of osteocytes f ound there. Osteocyte apoptosis may reduce the signaling available to initiate the replacement of damaged bone (73). This may explain why fracture risk increases as early as three months after GC use, even before significant bone loss has occurred (73). In response to this, th e Royal College of Physicians of London in recent years suggested using a T sc ore of -1.5 or lower as the treatment threshold for GC users (81). Despite these recommendations, it is estima ted that less than 15% of those on long-term steroid treatment also receive preventive medicat ion to prevent osteoporosis (16,22,89) and there still appears to be limited testi ng for bone loss in GC patients (22) There have been suggestions that preventive treatment should begin at the same time as GC therapy is initiated (89).

PAGE 49

49 Parathyroid Hormone Dual Nature of PTH*: Continuous versus Inte rmittent Administration Parathyroid hormone has a surprising, dual effect in mammals depending on whether delivery is continuous or intermitte nt. At least since the 1930s researchers have noted that PTH is catabolic when exposure was continuous (as it normally is in the body in response to low plasma calcium levels) but showed anabolic prop erties when administered intermittently (9093). The use of PTH in an anabolic role wa s not seriously pursued, however, until it became possible to manufacture syntheti c PTH. In the mid-1970s, recombinant techniques made it possible to sequence the amino acid fragment responsible for the horm ones anabolic effect (90). This anabolic capability resides in the first 34 amino acids on the N-terminal end which is why teriparatide is manufactured as PTH (1-34) (48,92,94). It has been suggested that at least some of the anabolic effects of intermittent PTH in bone are mediated through an IGF-I-dependent mechanis m, while the catabolic effects are mediated through gene expression that causes an increased ratio of RANKL to OP G (95,96). Teriparatide reverses the effects of GC on IGF-I expression in vitro, and th is may partially explain its effects in treating GC-induced bone loss (80). Parathyroid Hormone (PTH 1-84) Endogenous PTH is an 84-amino acid protein secr eted by the chief cells of the parathyroid glands when low serum calcium levels are de tected by calcium receptors on the parathyroid glands (97) or there are elevated levels of extr acellular phosphate (98). PTH 1-84 is a critical mediator of calcium homeostasis. Calciumsensing receptors on the surface cells of the parathyroid gland respond to minute-by-minute ch anges in serum calcium levels (48,94,99) and To avoid confusion, PTH produced by the parathyroid gl ands will be referred to as endogenous PTH or PTH 1-84 and recombinant parathyroid hormone will be referred to as teriparatide or PTH (1-34).

PAGE 50

50 maintain calcium balance through direct actions on bone and kidneys and indirectly through the gastrointestinal tract. When blood levels of cal cium are low, PTH stimulates bone resorption to liberate calcium stored in the bone matrix and enhances calcium reabsorption at the distal convoluted tubules of the kidney (94,98). PTH also regulates 1 alpha-hydroxylase activity in the kidney, facilitating the conversion of 25-hydr oxyvitamin D to 1,25 dihydroxyvitamin D in the kidney, which then acts on the GI tract to stim ulate increased absorption of calcium across the gut (94,99,100). In humans, the half-life of endogenous PTH in the blood is less than 3 minutes and it is metabolized by both the kidney (20-30%) and liver (60-70%) (90,94). This rapid metabolism means the availability of PTH is determined by the rate of secretion from the parathyroid glands. Endogenous PTH has both rapid and slow effects on bone. The rapid phase occurs within 30-90 minutes after exposure and is ch aracterized by increas ed osteoclast activity. A second, later phase, is associated with an in crease in both the number and ac tivity of osteoclasts (94,101). With continuous PTH exposure there is a decrea se in OPG mRNA and an increase in RANKL mRNA. While the actual mechanisms are not fu lly understood, PTH may exert its actions by activating a number of enzymes such as collag enase, lysosomal hydroxylases, acid phosphatases, H+, K+-adenosine triphosphatases, Na+/Ca+ exchange systems, cathepsin B, or cystein proteases. Continuous PTH exposure causes bone lining cells to retract from the bone surface as part of a calpain-dependent modification to the osteocyte cytoskeleton. This allows osteoclasts to attach to the bone surface where they th en initiate the process of bone resorption (94). PTH also inhibits osteoclast apoptosis, possibly by stim ulating the expression of RANKL and decreasing the expression of OPG by osteoblasts (102).

PAGE 51

51 Teriparatide (PTH 1-34) Teriparatide, a recombinant parat hyroid hormone marketed as FORTEOTM by Eli Lilly pharmaceutical company, is currently the only FDAapproved anabolic drug for osteoporosis. Unlike the anti-resorptive drugs, teriparatide st imulates bone formation, which contributes to increased bone mass, quality, and strength. Teri paratide is administer ed as a subcutaneous injection typically 20ug/day for 18-24 months in humans. The drug is not recommended for long-duration use since it caused an increased incidence of osteosarcoma in rats after long-term exposure to large doses of the drug (103-105). Teri paratide is manufactured from a strain of Escherichia coli modified by recombinant DNA technology (100,106). Teriparatide, which has a bioavailability of around 95% ( 90,100), reaches peak plasma con centration in about 30 minutes, then drops to virtually undete ctable levels in 3-4 hours (90,92 ,97,100). The systemic clearance of teriparatide is approximately 62 L/hour in wo men and 94 L/hour in men, which is greater than the rate of normal hepatic plasma flow, indicatin g both hepatic and kidney clearance similar to PTH (1-84) (100). Following extensive testing and clinical tria ls, teriparatide was approved by the FDA in November 2002 (100) for the treatment of postme nopausal women with osteoporosis who are at high risk for fracture and to increase bone mass in men with primary or hypogonadal osteoporosis who are at high risk for fractures (107). Daily inject ions of teriparatide increase bone mass, microarchitectural st ructure and bone streng th in mice, rats, rabbits, monkeys and humans (97). Studies have shown it provides a statistically significant increase in BMD at clinically important sites such as the lumbar spine (29). The increased bone turnover caused by teriparatide results in additional bone apposition on both periosteal and cancellous bone surfaces (97). The effects of teriparatide are less robust at the femoral neck.

PAGE 52

52 Mechanisms of Action PTH and teriparatide work through G Protei ns and the PTH Receptor 1 (PTH1R) has a similar affinity for both (90). PTH receptors are found predominantly on osteoblasts (but are also found in renal tubular cells) (48,90,94). Bi nding of PTH to the receptor activates adenylate cyclase and phospholipases A, C, and D and increas es intracellular levels of cAMP and calcium (90). Some think it may be the ability of PTH to stimulate both adenylate cyclase and phospholipase C that gives it its dual ana bolic and catabolic abilities (48). The primary effects of teriparatide on oste oclasts are indirect a nd are mediated by the drugs effects on osteoblasts. Increased bone formation following exposure to teriparatide occurs because of an increase in osteoblast numbers either through enhanced differentiation of pre-osteoblasts or an increase in the number of existing bone lini ng cells that differentiate into osteoblasts (48,90,108). Teriparatide also has an anti-apoptotic effect on osteoblasts, enabling them to secrete bone matrix for a longer period of time (48,92,100,109). Through these actions, the production of collagen-based bone matrix in creases, improving trabecular bone volume and connectivity (90,110). Teriparatide also acts on the cortical surface of cortical bone and increases its thickness without increasing porosity (90,109,111). Treatment with teriparatide has a variety of effects on osteoblas ts, including causing increased secretion of various growth factors, such as transforming growth factor (TGF), IGF I and II, IGF binding proteins, bon e morphogenic proteins (BMPs), and cytokines such as IL-1, Il-6 and M-CSF (48,92,95,112). MCSF, IL-1, RANKL, and TNF all enhance osteoclast survival (102). The net result is increased bone turnover. Howeve r, unlike the increased turnover seen in many disease conditions, the incr eased turnover associated with teriparatide favors bone formation over resorption. This is wh at causes teriparatide to have its anabolic effect on bone.

PAGE 53

53 Studies of Glucocorticoid-I nduced Bone Loss in Mice Validity of the Mouse as a Model of Glucocorticoid-Induced Bone Loss The mouse genome has now been sequenced, maki ng this animal an even more attractive model for scientific research. The mouse has show n its value in studies of bone loss due to aging and sex steroid alterations (3) and some believe it is the preferred rodent model for the study of bone loss due to GC exposure (3-5). This is be cause some researchers believe the rat may be resistant to the deleterious effects on bone asso ciated with GC exposure and therefore may not represent a good model for this specific condition (36-38). Commonly used mouse strains to assess the effects of GCs include the Swiss Webster (5), the C 57B1 (4) and the Balb/C (3) (see table 2). Studies to date have used techni ques such as histomorphometry, microCT, serum biochemistry, and DXA to measure bone responses in mice exposed to GCs (2-5). Most studies have reported that GC treatment induces greate r axial than appendicula r bone loss (5) without significant weight loss or hypogona dism (2,3). A summary of GC -induced bone loss studies using mice listed in Appendix B. Glucocorticoid-Induced Bone Loss in Mice Glucocorticoid drugs have been shown to affect mouse bone metabolism during both in vitro and in vivo experiments. In cell culture studies, the number of osteoclast precursors following treatment with prednisolone for 4 or 10 da ys decreased significantly after just 4 days. Osteoblast precursors also decreased, but only after 10 days of treatment. Despite fewer precursor cells, prednisolone exposure resulted in an 81% increase in osteoclast numbers and a 20-fold increase in the ratio of osteoclasts to osteoblasts perhaps reflecting decreased osteoclast apoptosis (4). In adult Swiss Webster mice, glucocorticoid s affect osteoblasts and osteocytes at a threshold dose of 1.4 mg/kg body weight (5). Some researchers have used this threshold dose (2)

PAGE 54

54 while others have used a dose of 2.1 mg/kg body weight (4,5). At this higher level of treatment, researchers have found increased oste oclast survival in as few as 10 days (4). MicroCT analysis following administration of low levels of GC ( 1.2 mg/kg) only found signi ficant differences in BV, while there were no significant differences either in Tb.N or Tb.Th (2). Bone changes were generally dose and duration dependent. In one st udy the greatest changes were detected when 10 mg/kg was administered ove r a 21-day period (3). Histomorphometric analysis of the bones ha s yielded mixed results with some studies finding significant decreases in trabecular thickness while ot her studies found no significant difference in this measure (2-5). In these st udies, histomorphometric analysis of kinetic measures were more consistent and showed significant decreases in mineralizing surface, mineral apposition rate, and bone formation rate (2-5). Studies have also demonstrated a preferential bone loss in the ax ial skeleton (5) and that spin al BMD decreases in a dosedependent manner (3-5). Dynamic measures, such as mineralizing surf ace (MS), mineral apposition rate (MAR), and bone formation rate (BFR/BS) which are only available using histomorphometry, showed changes in Swiss Webster mice at dosages > 1.2 mg/kg (2,4,5) while it t ook dosages of 10mg/kg to elicit changes in Balb/C mice (3). Teriparatide Treatment in Mice Teriparatide use has resulted in increased bone density and strength when used in mice (2,4,5,7,8). There have been studie s treating intact and ovariectom ized mice with teriparatide (see table 3-2), though none of th ese studies examined the eff ectiveness of teriparatide in preventing or reversing GC-induced bone loss in mice. Studies most commonly used the C57BL/6 (8,35,45), CBA-1 (113), and Swiss Webster st rains (7) of mice. In the strains of mice

PAGE 55

55 tested, teriparatide had the grea test anabolic effect in the femur and was dose and duration dependent (7,8,45). One study found a subcutaneous dose of 40g/ kg/day 5 days/week increased BMD within 1-2 weeks in the tibia and within 7 weeks in th e vertebrae, suggesting site specific differences occur (45). This finding differs from anothe r study which found the earliest effects in the vertebrae (8). Research ers have reported increased bone mass in cortical as well as cancellous bone, although increases in cortical bone were found primarily in long bones, possibly reflecting a response to mechanical loading patterns (8). Significant effects from teriparatide seem to depend on the presence of existing trabeculae and may be hampered in areas that have suffered severe disruption of trabeculae (8 ). A summary of studies usi ng PTH in mice is in Appendix B Further Considerations We do not fully understand the mechanisms governing GC-induced osteoporosis nor how they effect bone mass and micr oarchitectural structure. Th is study seeks to further our understanding of these processes in mice as a prelude to human studi es. To date, there has only been one study examining the efficacy of te riparatide in treating glucocorticoid-induced osteoporosis. In this study, postmenopausal women on long-term GC therapy and hormone replacement therapy (HRT) also received teripara tide. The addition of te riparatide resulted in significant bone density increases in the axial skeleton (6). No study of teriparatide alone in glucocorticoid-induced osteoporosis in a human or murine model has been done to date.

PAGE 56

56 CHAPTER 4 RESULTS Measurement Design During this experiment, 70 7-month old male Swiss Webster mice were randomized into 7 groups which received prednisolone teriparatide, or vehicle in an attempt to characterize the effects of teriparatide on bone in glucocorticoid-treated mice. Gr oups are identified in Table 4-1. At the end of the study, mice were euthanized an d both femurs and lumbar vertebrae 2 and 3 (LV 2, LV3) were removed and analyzed using hist omorphometry or microCT. Histomorphometric techniques were used on the right femur and L3 to measure the static parameters of BV/TV, Tb.N, Tb.Wi, Tb.Sp and dynamic parameters of MS, MAR, and BFR/BS. MicroCT measures included BV/TV, Tb.N, Tb.Th, and Tb.Sp in the le ft femur and L2. Cortical BV/TV was also measured in the mid-shaft of the left femu rs. Table values are reported as mean + standard deviation and figure values are reported as mean + standard error. Table 4-1. Experimental Groups an d Description of Treatments. Group n = Treatment (6 days/week) BSL CNTL 10 Baseline Control 8 WK CNTL 10 8-Week Control. Received GC-vehicle and PTHvehicle for 8 weeks GC4/SAC 10 Received GC and PTH-vehicle for 4 weeks and were then sacrificed GC4/RECV 10 Received GC and PTH-vehicle for 4 weeks and then received GC-Vehicle and PTH-Vehicle for 4 weeks to allow for natural recovery GC8 10 Received GC and PTH-vehicle for 8 weeks GC4/GC-PTH4 10 Received GC and PTH-vehicle for 4 weeks and then GC and PTH for 4 weeks GC-PTH8 10 Received GC and PTH for 8 weeks GC = prednisolone, 2.1 mg/kg/day; GC-vehicle = sterile saline; PTH = teriparatide 40ug/kg/day; PTH-vehicle = 2% acidified mouse serum.

PAGE 57

57 Anthropomorphic Measures Animals were weighed at study entry, weekly, an d prior to sacrifice. Weight changes were used to monitor animal health and adjust dosages of study drugs. Weight data are presented in Table 4-2. There were no statisti cal differences in the average we ight between the groups either at the start or end of the experiment. Table 4-2. Mean Animal Weights by Group. GROUP Start Wt (g)End Wt (g)Diff (g) BSL CNTL (n =10) 35.8 + 2.7 n/a 8 WK CNTL (n =10) 35.2 + 3.436.1 + 2.6+ 0.9 GC4/SAC (n = 10) 35.3 + 3.734.0 + 4.21.3 GC4/RECV (n = 10) 35.7 + 2.536.9 + 3.5+ 1.2 GC 8 (n = 10) 35.2 + 2.334.4 + 3.70.8 GC4/GC-PTH4 ( n = 10) 35.6 + 1.835.1 + 2.80.5 GC8/PTH8 ( n = 10) 35.1 + 4.035.1 + 4.2 0.0 Values are expressed as mean + standard deviation. Previous studies indicate 7-month old Swiss Webster mice have reached skeletal maturity and have ceased longitudinal bone growth (5). To verify cessation of long bone growth, each animals left femur was measured and the resu lts are presented in Table 4-3. There were no significant differences in fe mur length between the groups. Table 4-3. Mean Femur Lengths by Group. BSL CNT (n = 10) 8 WK CNTL (n = 10) GC4/ SAC ( n = 10) GC4/ RECV ( n = 10) GC 8 ( n = 10) GC4/ GC-PTH4 ( n = 10) GC8/ PTH8 (n= 10) Femur Length(mm) 15.5 + 0.4 15.4 + 0.3 15.3 + 0.4 15.6 + 0.3 15.5 + 0.2 15.6 + 0.5 15.4 + 0.4 Results are reported as mean + standard deviation.

PAGE 58

58 Bone Measures Histomorphometric analysis was performed on th e right femurs of 67 animals and the third lumbar vertebra (L3) of 70 animals. One femu r was damaged during tissue harvesting and could not be sectioned and the other tw o were sectioned but could not be analyzed. Static and dynamic parameters of bone change were measured or derived using methods described in Chapter 3. Measured parameters using histomorphometry included BV/TV, Ob.S Oc.S, MS and MAR. Once BV/TV was determined, values for Tb.N, Tb.Wi*, and Tb.Sp were derived using calculations described elsewhere (51). On ce MS and MAR were measured, BFR/BS was calculated as the product of thes e two. MicroCt was performed on 70 intact femurs and 70 intact second lumbar vertebrae (L2). Static parame ters of bone change measured using microCT include BV/TV, Tb.N, Tb.Th*, and Tb.Sp. Data are presented as the measured or derive d values for each parameter. There are also tables showing percent change between groups that reached statistical signif icance. These tables show the percent change compared to the group in the first column of the table. Percent change is reported for all signifi cant group interactions but only significant differences (p < 0.05) or trends among comparable groups will be discusse d. To simplify data presentation, a separate table is provided within the chapter for each measured or derived parameter. Appendix A, however, contains comprehensiv e tables for static, dynamic and cellular data by bone for each measurement technique (histomo rphometry and microCT). Measurements of the Lumbar Vertebrae Histomorphometry data for L3 are summarized in Table 4-4. Lumbar vertebrae L2 was used for microCT and the results are shown in Table 4-5. Histomorphometry analyzes a two-dimensional sample so the distance across trabeculae will be reported as Tb.Wi; microCT analyzes samples in three dimensions so the distance across the trabeculae is reported as Tb.Th.

PAGE 59

59 Table 4-4. Summary of Histomorphom etric Analysis of LV3 by Group. BSL CNTL (n = 10) 8 WKCNTL (n = 10) GC4/SAC (n = 10) GC 4/RECV (n = 10) GC 8 (n = 10) GC4/PTH4 (n = 10) GC-PTH8 (n = 10) BV/TV (%) 12.7 + 3.3 8.8+ 3.2 10.9 + 3.0 9.2+ 3.4 8.7 + 2.4 11.0 + 3.1 12.5 + 2.1 Tb.N (1/mm) 4.9 + 0.8 4.1+ 1.3 4.7 + 0.8 4.3 + 0.9 4.5 + 1.1 4.1 + 0.7 4.9 + 0.6 Tb.Wi ( m ) 31.0 + 4.6 25.3+ 5.0 27.5 + 3.8 25.1+ 5.1 23.2 + 3.0 31.5 + 4.3 30.9 + 4.9 Tb.Sp ( m ) 185.6 + 38.4 242.1+ 82.4 196.7 + 42.4 223.0 + 6.4 214.5 + 6.1 220.7 + 39.6 182.3 + 21.0 Results are reported as mean + standard deviation. Table 4-5. Summary of MicroCT Analysis of LV2 by Group. BSLCNTL (n= 10) 8 WK CNTL (n = 10) GC4/SAC (n = 10) GC4/RECV (n = 10) GC 8 (n = 10) GC4/PTH4 (n = 10) GC-PTH8 (n = 10) BV/TV (%) 19.0 + 3.6 16.1 + 3.6 17.4 + 4.3 15.0 + 3.8 19.0 + 4.6 18.8 + 4.8 22.0 + 4.4 Tb.N (1/mm) 4.2 + 0.3 3.9 + 0.5 3.9 + 0.5 3.8 + 0.5 4.4 + 0.8 3.8 + 0.5 4.04 + 0.4 Tb.Th ( m ) 48.7 + 2.8 47.5 + 3.8 47.1 + 1.9 45.9 + 3.0 46.4 + 2.2 52.1 + 4.2 54.7 + 4.9 Tb.Sp ( m ) 236 + 19 259 + 40 261 + 33 266 + 35 232 + 50 265 + 35 245 + 29 Results are reported as mean + standard deviation. Bone Volume/Total Volume Bone volume data based on histomorphometry are presented in Table 4.4, Table 4-6 and Figure 4-1. Bone volume data based on microCT are presented in Table 4-5, Table 4-7, and Figure 4-2. Bone volume was lowe r in older animals and higher in animals treated with PTH as shown in Figure 4-1 and Figure 4-2. Histomor phometry indicated BV/TV was significantly lower in the 8 WK CNTL (-30.7%, p = 0.03) gr oup compared to BSL CNTL. Significant increases in BV/TV were seen in animals treated with PTH. Animals treated with PTH for 8 weeks had a higher BV/TV than both the 8 WK CNTL (+42%, p = 0.01) and GC8 (+ 43.7%, p = 0.01) groups and a 35.9% (p = 0.03) higher BV/TV th an animals in GC4/RECV. Animals in the GC4/GC-PTH4 group had a 26.4% (p = 0.03) high er BV/TV than GC8 animals. MicroCT analysis found significantly hi gher BV/TV in GC-PTH8 compar ed to 8 Wk CNTL (+36.6%, p =

PAGE 60

60 0.01), GC4/SAC (+26.4%, p = 0.04), and GC4/RECV (+46.7%, p = 0.00). There was no bone loss in GC8 compared to 8WK CNTL us ing either histomorphometry (8.7 + 2.4 vs. 8.8 + 3.2) or microCT (19.0 + 4.6 vs. 16.1 + 3.6). BV/TV did not increase after GC treatment ended. Table 4-6. Significant Change s in Lumbar Vertebra L3 Bone Volume by Group using Histomorphometry. GROUP 8WK CNTL GC4/RECV GC8 GC4/GCPTH4 GC-PTH8 BSL CNTL 30.7% (p = 0.03) 27.7% (p = 0.03) 31.5% (p=0.02) 8 WK CNTL + 42.0% (p = 0.01) GC4/RECV + 35.9% (p = 0.03) GC 8 + 26.4% (p = 0.03) + 43.7% (p = .01) Results are percent change compar ed to groups on the left. Figure 41. Lumbar Vertebra L3 B one Volume/Total Volume by Group using Histomorphometry. a = significant comp ared to BSL CNTL; b = significant compared to 8 WK CNTL; c = significan t compared to GC4/RECV; d = significant compared to GC8. Results are percent ch ange compared to groups on the left. Results are reported as mean + standard error. Bone Volume/Total Volume (BV/TV) 0 2 4 6 8 10 12 14 16 BSL CNTL 8WK CNTL GC4/SAC GC4/RECV GC8 GC4/GCPTH4 GC-PTH8% a a a b,c,d d

PAGE 61

61 Table 4-7. Significant Changes in Lumbar Vertebra L2 Bone Vo lume by Group using MicroCT. Group GC4/RECV GC-PTH8 BSL CNTL 21.1% (p = 0.03) 8 WK CNTL + 36.6% (p = 0.01) GC4/SAC + 26.4% (p = 0.04) GC4/RECV + 46.7% (p = 0.00) GC 8 21.1% (p = 0.04) Results are shown as percent change co mpared to groups on the left. Figure 4-2. Lumbar Vertebra L2 Bone Volume/Total Volume by Group using MicroCT. a = significant compared to BSL CNTL; b = si gnificant compared to 8 WK CNTL; c = significant compared to GC 4/SAC; d = significant compared to GC4/RECV; e = significant compared to GC8. Re sults are reported as mean + standard error. Trabecular Number Data for trabecular number are presented in Ta ble 4-4, and Figure 43 (histomorphometry) and Table 4-5 and Figure 4-4 (microCT). Acco rding to histomorphometr y, the only significant difference in trabecular number in L3 among treatment groups was between the GC4/GC-PTH4 Bone Volume/Total Volume (BV/TV) 0 5 10 15 20 25 BSL CNTL 8WK CNTL GC4/ SAC GC4/ RECV GC8 GC4/GCPTH4 GC-PTH8% a,e b,c,d

PAGE 62

62 and GC-PTH8, where the latter group had a hi gher (19.5%, p = 0.02) Tb.N. MicroCT, on the other hand, showed only a non-sign ificant 6.3% difference in thes e groups (Appendix A). There was, however, a significant difference based on microCT between GC4/RECV (-9.5%, p = 0.04) and GC4/GC-PTH4 (-9.5%, p = 0.02) compared to BSL CNTL. Figure 4-3. Lumbar Vertebra L3 Trabecular Number by Group using Histomorphometry. a = significant compared to GC4/GC-PTH4. Results are reported as mean + standard error. Figure 4-4. Lumbar Vertebra L2 Trabecular Number by Group us ing MicroCT. a = significant compared to BSL CNTL. Results are reported as mean + standard error. Trabecular Number (Tb.N) 0 1 2 3 4 5 6 7 BSL CNTL 8WK CNTL GC4/ SAC GC4/ RECV GC8 GC4/GCPTH4 GC-PTH8 1/mm a Trabecular Number (Tb.N) 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 BSL CNTL 8WK CNTL GC4/ SAC GC4/ RECV GC8 GC4/GCPTH4 GC-PTH8 1/mm aa

PAGE 63

63 Trabecular Width/Trabecular Thickness Trabecular Width/Trabecular Thickness was a measure of the mean distance across the trabeculae and changes reflected the effects of prednisolone, teriparatide, or both on this distance. Data for Tb.Wi based on histomor phometry are found in Table 4-4, Table 4-8 and Figure 4-5, while microCT data concerning Tb.T h are presented in Table 4-5, Table 4-9, and Figure 4-6. As shown in Figure 4-5, histomorphometry generally detected higher Tb.Wi in PTHtreated animals and lower Tb.Wi in older anim als but no significant difference based on GC treatment in comparably aged animals. There was a significantly lowe r Tb.Wi between the BSL CNTL group and animals in the 8 WK CNTL (-18.4%, p = 0.02). Histomorphome try and microCT each found significant differences between both of the PTH groups a nd 8 WK CNTL, GC4/RECV and GC8. There was also a higher Tb.Wi in GC4/GC-PTH4 ( 24.5%, p = 0.01) and GC-P TH8 ( 22.1%, p = 0.01) groups compared with comparably aged animals (8 WK CNTL). Additionally, animals treated with PTH for either four or eight weeks ha d a 35.8% (p = 0.00) and 33.2% (p = 0.00) higher Tb.Wi respectively compared to GC8 animals. There was no significant difference in Tb.Th between GC4/GC-PTH4 and GC-PTH8. Similar to histomorphometric findings, micr oCT showed significan tly higher Tb.Th in animals treated with PTH. Animals in th e GC4/GC-PTH4 group had a greater Tb.Th than animals in the 8 WK CNTL (+9.7%, p = 0.04), GC4/SAC (+10.6%, p = 0.01), GC4/RECV (+13.5%, p = 0.00) and GC8 (+12.3%, (p=0.00) groups. Animals tr eated with PTH for eight weeks showed a higher Tb.Th than 8 WK CNTL (+15.2%, p = 0.00), GC4/SAC (+16.1%, p = 0.00), GC4/RECV (+19.2%, p = 0.00), and GC 8 (+17.9%, p = 0.00). Again, there was no significant difference in Tb.Th between the GC4/GC-PTH4 and GC-PTH8 groups.

PAGE 64

64 Table 4-8. Significant Change s in Lumbar Vertebra L3 Trabecular Width by Group using Histomorphometry. Group 8WK CNTL GC4/SACGC4/RECVGC8 GC4/GCPTH4 GC-PTH8 BSL CNTL 18.4% (p = 0.02) 19.0 (p=0.02) 25.1 (p= 0.00) 8 WK CNTL + 24.5 % (p =0.01) + 22.1 (p = 0.01) GC4/SAC + 14.5% (p = 0.03) GC4/RECV + 25.5% (p = 0.00) + 23.1% (p = 0.01) GC 8 + 18.5% (p = 0.02) + 35.8% (p = 0.00) + 33.2% (p = 0.00) Results are shown as percent change co mpared to groups on the left. Trabecular Width (Tb Wi) 0 5 10 15 20 25 30 35 40BSL CNTL8WK CNTL GC4/SACGC4/RECVGC8GC/GCPTH4 GC-PTH8micrometer aa a b, d, e b, c, d, e Figure 4-5. Lumbar Vertebra L3 Trabecular Width by Group using Histomorphometry. a = significant compared to BSL CNTL; b = si gnificant compared to 8 WK CNTL; c = significant compared to GC 4/SAC; d = significant compared to GC4/RECV; e = significant compared to GC8. Re sults are reported as mean + standard error.

PAGE 65

65 Table 4-9. Significant Changes in Lumbar Vertebra L2 Trabecular Thickness by Group using MicroCT. Group GC4/RECV GC4/GC-PTH4 GC-PTH8 BSL CNTL 5.7% (p = 0.02) + 12.3% (p = 0.01) 8 WK CNTL + 9.7% (p = 0.04) + 15.2% (p = 0.00) GC4/SAC + 10.6% (p = 0.01) + 16.1% (p = 0.00) GC4/RECV + 13.5% (p = 0.00) + 19.2% (p = 0.00) GC 8 + 12.3% (p = 0.00) + 17.9% (p = 0.00) Results are shown as percent change co mpared to groups on the left. Figure 4-6. Lumbar Vertebra L2 Trabecular Th ickness by Group using MicroCT. a = significant compared to BSL CNTL; b = significant compared to 8 WK CNTL; c = significant compared to GC4/SAC; d = significant co mpared to GC4/RECV; e = significant compared to GC8. Results are reported as mean + standard error. Trabecular Thickness (Tb.Th) 0 10 20 30 40 50 60 70 80 BSL CNTL 8WK CNTL GC4/ SAC GC4/ RECV GC8 GC/GCPTH4 GC-PTH8micrometers a a,b,c d,e b,c d,e

PAGE 66

66 Trabecular Separation Trabecular separation data are shown in Tabl e 4-4 and Figure 4-7 (histomorphometry) and Table 4-5 and Figure 4-8 (microCT). Trabecular Separation is a reflection of the mean distance between trabeculae within the region of interest. This distance tends to increase when resorption is higher and decrease with incr eased bone formation. MicroCT re sults typically showed greater trabecular separation for each group than di d histomorphometric analysis, although the differences were not significant. There were no significant differen ces in Tb.Sp related to age or GC exposure based on microCT or histomorphomet ry. The only significant differences in trabecular separation in L3 (histomorphometry) were found in PTH-treated animals. The GCPTH8 group had a significantly lower (-24.7%, p = 0.03) Tb.Sp compared with the 8 WK CNTL and GC4/GC-PTH4 (-17.4%, p = 0.01) groups (Appe ndix A). MicroCT indicated significant differences between BSL CNTL and GC4/RE CV (+12.7%, p = 0.03) and GC-PTH8 (+12.3%, p = 0.04). Figure 4-7. Lumbar Vertebra L3 Trabecular Se paration by Group using Histomorphometry. a = significant compared to 8 WK CNTL; b = significant compared to GC4/GC-PTH4. Results are reported as mean + standard error. Trabecular Separation (Tb.Sp) 0 50 100 150 200 250 300 BSL CNTL 8WK CNTL GC4/ SAC GC4/ RECV GC8 GC4/GCPTH4 GC-PTH8micrometers a,b

PAGE 67

67 Figure 4-8. Lumbar Vertebra L2 Trabecula r Separation by Group using MicroCT. a = significant compared to BSL CNTL. Results are reported as mean + standard error. MicroCT methods showed significant changes in Tb. Sp only in relation to the BSL CNTL group. Histomorphometry did not find changes in any group compared to the BSL CNTL group but did find significant differences in animals treated with PT H, but only after 8 weeks of treatment. Measurements of the Distal Femur The right femurs of 67 animals used for this study were harvested for histomorphometric analysis while 70 left femurs from the same an imals were analyzed using microCT. Three bones could not be used for histomorphometric analysis for reasons previously described. The femurs were harvested from each animal as described in Chapter 2. From each right femur, two 4 m thick sections were stained and used to measur e or derive static bone measurements including BV/TV, Tb.N, Tb.Wi, and Tb.Sp. The results of that analysis are summarized in Table 4-10. Femurs used for microCT were scanned intact and the resulting measur es of BV/TV, Tb.N, Tb.Th, and Tb.Sp are presented in Table 4-11. Trabecular Separation (Tb.Sp) 0 50 100 150 200 250 300 350 BSL CNTL 8WK CNTL GC4/ SAC GC4/ RECV GC8 GC4/GC PTH4 GC-PTH8 micrometers a a

PAGE 68

68 Table 4-10. Summary of Histom orphometric Analysis of the Distal Femur by Group. BSL CNTL (n = 10) 8 WK CNTL (n = 9) GC4/SAC (n = 10) GC4/RECV (n -= 10) GC 8 (n=9) GC4/GC-PTH4 (n=9) GC-PTH8 (n=10) BV/TV (%) 4.8 + 2.3 2.6 + 1.9 4.8 + 2.3 2.3 + 2.1 3.1 + 1.6 4.5 + 1.7 5.2 + 2.9 Tb.N(1/mm) 2.0 + 0.9 1.2 + 0.8 1.8 + 0.9 1.2 + 0.9 1.6 + 0.8 1.9 + 0.7 2.2 + 0.9 Tb.Wi( m ) 28.2 + 5.9 25.4 + 7.2 26.1 + 5.9 21.9 + 5.5 23.2 + 6.1 27.9 + 4.3 28.6 + 6.1 Tb.Sp ( m ) 540 + 225 1366+ 1146 694 + 428 1497 + 1812 904 + 916 551.6 + 195 554 + 359 Oc.S (%) 0.6 + 0.2 1.1 + 1.1 0.4 + 0.3 1.3 + 0.8 0.9 + 0.7 0.8 + 0.6 1.7 + 1.2 Ob.S (%) 7.9 + 13.2 7.2 + 3.5 1.1 + 1.2 8.7 + 8.9 2.1 + 1.5 13.6 + 12.0 20.0 + 7.8 MS (%) 6.7 + 5.4 3.1 + 3.2 1.1 + 1.6 5.9 + 3.5 1.6 + 1.8 14.2 + 5.3 16.8 + 8.7 MAR (m/d) 0.9 + 0.2 0.6 + 0.2 0.6 + 0.2 0.7 + .1 0.6 + 0.2 0.8 + 0.1 0.8 + 0.2 BFR/BS (um3/um2/d) 6.4 + 6.0 2.2 + 3.0 0.6 + 1.0 4.2 + 2.7 1.0 + 1.1 10.8 + 5.7 14.8 + 12.8 Results are reported as mean + standard deviation. Table 4-11. Summary of Mi croCT Analysis of the Distal Femur by Group. BSL CNTL (n = 10) 8WK CNTL (n = 10) GC4/SAC (n = 10) GC4/RECV (n = 10) GC 8 (n = 10) GC4/PTH4 (n = 10) GC-PTH8 (n = 10) BV/TV (%) 9.9 + 3.1 6.2 + 2.1 8.5 + 3.4 6.0 + 3.8 8.7 + 2.6 7.9 + 2.9 11.5 + 3.1 Tb.N(1/mm) 2.7 + 0.63 2.1 + 0.45 2.5 + 0.57 2.2 + 0.78 2.8 + 0.75 2.5 + 0.54 2.8 + 0.47 Tb.Th ( m ) 60.7 + 5.2 64.5 + 6.3 60.1 + 2.8 59.9 + 6.4 56.3 + 8.5 63.9 + 5.0 70.6 + 7.2 Tb.Sp ( m ) 383 + 91 485 + 113 413 + 99 500 + 158 391 + 154 416 + 108 361 + 60 Results are mean + standard deviation. Bone Volume Bone volume data based on histomorphometry ar e presented in Table 4-10, Table 4-12 and Figure 4-9 while microCT data are presente d in Table 4-11, Table 4-13, and Figure 4-10. Histomorphometry and microCT detected the same trends in BV/TV in the distal femur. Bone volume tended to be lower with age and higher with exposure to PTH. There was a lower BV/TV in 8Wk CNTL compared to BSL CNTL in histomorphometry and microCT. GCPTH treatment resulted in hi gher BV/TV compared to 8 WK CNTL control and GC-treated groups. Compared to GC/RECV, both groups receiving PTH had signifi cantly increased bone volume

PAGE 69

69 Table 4-12. Significant Cha nges in Distal Femur Bone Volume by Group using Histomorphometry. Group 8WK CNTL GC4/RECV GC4/GC-PTH4 GC-PTH8 BSL CNTL 45.8% (p = 0.03) 52.1% (p = 0.01) 8WK CNTL +100.0% (p = 0.02) GC4/RECV + 95.6% (p = 0.01) + 126.1% (p = 0.01) Results are shown as percent change co mpared to groups on the left. Bone Volume/Total Volume (BV/TV)0 1 2 3 4 5 6 7 BSL CNTL8WK CNTL GC4/SACGC4/RECVGC8GC4/GCPTH4 GC-PTH8% a b, c c a Figure 4-9. Distal Femur Bone Volume/Total Volume by Group using Histomorphometry. a = significant compared to BSL CNTL; b = si gnificant compared to 8 WK CNTL; c = significant compared to GC4/RECV. Results are reported as mean + standard error. There was only a slightly different BV/T V between GC4/SAC animals and the GC4/GCPTH4 and GC-PTH8 groups and these differences di d not reach statistical significance. The GCPTH8 group had a higher BV/TV than GC4/GC-PTH4 (45.6%, p = 0.02) based on microCT. MicroCT and histomorphometric comparisons of PTH groups with the 8 WK CNTL showed BV/TV was higher in GC4/GC-PTH4 and the GC-P TH8 group, despite the fact these groups also received GC.

PAGE 70

70 Table 4-13. Significant Changes in Distal Femur Bone Volume by Group using MicroCT. Group 8WK CNTL GC4/RECV GC8 GC-PTH8 BSL CNTL 37.4% (p = 0.00) 39.4% (p = 0.00) 8 WK CNTL + 40.3% ( p = 0.03) + 85.5% (p = 0.00) GC4/RECV + 45.0% ( p =0.02) + 91.7% (p = 0.00) GC 8 + 32.3% (p = 0. 0.04 GC4/GC-PTH4 + 45.6% ( p = 0.02) Results are shown as percent change compared to groups on the left. Figure 4-10. Distal Femur Bone Volume/T otal Volume by Group using MicroCT. a = significant compared to BSL CNTL; b = si gnificant compared to 8 WK CNTL; c = significant compared to GC4/RE CV; d = significant compared to GC8; e = significant compared to GC4/GC-PTH4. Results are reported as mean + standard error. BoneVolume/Total Volume (BV/TV) 0 2 4 6 8 10 12 14 16 BSL CNTL 8WK CNTL GC4/ SAC GC4/ RECV GC8 GC4/GCPTH4 GC-PTH8 % a b,c,d,e a b,c

PAGE 71

71 Trabecular Number Data for trabecular number based on histom orphometry are found in Table 4-10, Table 414, and Figure 4-11. MicroCT data on this para meter are found in Table 4-11, Table 4-15, and Figure 4-12. Histomorphometry and microCT det ected similar patterns of change; where Tb.N was lower in older animals and higher in animals treated with GC or GC-PTH. Table 4-14. Significant Cha nges in Distal Femur Trab ecular Number by Group using Histomorphometry. Group 8WK CNTL GC4/RECV GC-PTH8 BSL CNTL 40.0% (p = 0.05) 40.0% (p = 0.01) 8 WK CNTL + 83.3% (p = 0.03) GC4/RECV + 83.3% (p = 0.03) Results are shown as percent change compared to groups on the left. Figure 4-11. Distal Femur Trabecular Numb er by Group using Histomorphometry. a = significant compared to BSL CNTL; b = si gnificant compared to 8 WK CNTL; c = significant compared to GC4/RECV. Results are reported as mean + standard error. Trabecular Number (Tb.N) 0 0.5 1 1.5 2 2.5 3 BSL CNTL 8WK CNTL GC4/ SAC GC4/ RECV GC8 GC4/ GC-PTH4 GCPTH8 1/mm a a b,c

PAGE 72

72 Table 4-15. Significant Changes in Distal Femur Trabecular Numb er by Group using MicroCT. Group 8WK CNTL GC8 GC-PTH8 BSL CNTL 22.2% ( p=0.05) 8 WK CNTL + 33.3% (p = 0.03) + 33.3% (p = 0.00) GC4/RECV + 27.3% (p =0.01) Results are shown as percent change co mpared to groups on the left. Figure 4-12. Distal Femur Trabecular Number by Group using MicroCT. a = significant compared to BSL CNTL; b = significant co mpared to 8 WK CNTL; c = significant compared to GC4/RECV. Resu lts are reported as mean + standard error. Trabecular Thickness/Trabecular Width Data for Tb.Wi in the distal femur based on histomorphometry are found in Table 4-10, Table 4-16, and Figure 4-13. Micr oCT data for distal femur Tb.Th are found in Table 4-11, Table 4-17, and Figure 4-14. Hi stomorphometry and microCT detected the same general patterns in Tb.Th/Tb.Wi in the distal femur. Pa rameters of Tb.Wi and Tb.Th tended to be lower in response to GC treatment and higher with PTH treatment. There was no significant age effect Trabecular Number (Tb.N) 0 0.5 1 1.5 2 2.5 3 3.5 BSL CNTL 8WK CNTL GC4/ SAC GC4/ RECV GC8 GC4/GCPTH4 GC-PTH8 1/mm a b b,c

PAGE 73

73 apparent in Tb.Wi/Tb.Th as shown in Figure 413 and Figure 4-14. Both histomorphometry and microCT found significant differences between PTH-treated animals and the 8 WK CNTL and GC4/RECV groups. Additionally, there was a trend toward decreased Tb.N in GC4/SAC compared to GC4/RECV in both histomorphometry. Table 4-16. Significant Cha nges in Distal Femur Trab ecular Width by Group using Histomorphometry. Group GC4/RECV GC8 GC4/GC-PTH4 GC-PTH8 BSL CNTL 22.3% (p = 0.04) 17.7% (p = 0.05) GC4/RECV + 27.4 (p = 0.05) + 30.6% (p = 0.02) GC 8 + 20.3% (p =0.04) + 23.3% (p = 0.03) Results are shown as percent change co mpared to groups on the left. Figure 4-13. Distal Femur Trabecular Th ickness by Group using Histomorphometry. a = significant compared to BSL CNTL; b = si gnificant compared to GC4/RECV; c = significant compared to GC8. Re sults are reported as mean + standard error. Trabecular Width (Tb.Wi) 0 5 10 15 20 25 30 35BSL CNTL 8WK CNTL GC4/ SAC GC4/ RECV GC8 GC4/ GC-PTH4 GC-PTH8 micrometers a a b,c b,c

PAGE 74

74 Table 4-17. Significant Change s in Distal Femur Trabecular Thickness by Group using MicroCT. Group GC8 GC4/GC-PTH4GC-PTH8 BSL CNTL + 16.3% (p = 0.01) 8 WK CNTL 12.7% (p = 0.04) GC4/SAC + 6.3% ( p = 0.05) + 17.5% (p = 0.00) GC4/RECV + 17.9% (p = 0.00) GC 8 + 13.5% (p = 0.05) + 25.4% (p = 0.00) GC4/GC-PTH4 + 10.5% (p = 0.05) Results are shown as percent change compared to groups on the left. Figure 4-14. Distal Femur Trabecular Thic kness by Group using MicroCT. a = significant compared to BSL CNTL; b = significant compared to 8WK CNTL; c = significant compared to GC4/SAC; d = significant co mpared to GC4/RECV; e = significant compared to GC8; f = significant compared to GC4/GC-PTH4. Results are reported as mean + standard error. Trabecular Thickness (Tb.Th) 0 10 20 30 40 50 60 70 80 90 100 BSL CNTL 8WK CNTL GC4/ SAC GC4/ RECV GC8 GC/GCPTH4 GC-PTH8 micrometers a,c, d,e, f c,e b

PAGE 75

75 Trabecular Separation Data for trabecular separati on based on histomorphometry ar e found in Table 4-10, Table 4-18, and Figure 4-15. Data from microCT for Tb.Sp are found in Table 4-11, Table 4-19, and Figure 4-16. In general, higher Tb.Sp was seen in older animals and both GC and GC-PTH tended to decrease this parameter as shown in Figure 4-15 and Figure 4-16. Compared to the BSL CNTL group, Tb.Sp was hi gher (+153.0%, p = 0.05) in the 8WK CNTL group based on histomorphometry. Table 4-18. Significant Change s in Distal Femur Trabecu lar Separation by Group using Histomorphometry. Group 8WK CNTL GC4/RECV GC4/GC-PTH4 GC-PTH8 BSL CNTL +153.0% (p = 0.05) + 177.2% (p = 0.01) 8 WK CNTL 59.4% (p = 0.03) GC4/RECV 63.3% (p = 0.01) 63.0% (p = 0.03) Results are shown as percent change co mpared to groups on the left. Figure 4-15. Distal Femur Trabecular Sepa ration by Group using Histomorphometry. a = significant compared to BSL CNTL; b = si gnificant compared to 8 WK CNTL; c = significant compared to GC4/RECV. Results are reported as mean + standard error. Trabecular Separation (Tb.Sp) 0 500 1000 1500 2000 2500 BSL CNTL 8WK CNTL GC4/ SAC GC4/ RECV GC8 GC4/GCPTH4 GC-PTH8 micrometers a a b, c c

PAGE 76

76 Table 4-19. Significant Change s in Distal Femur Trabecu lar Separation by Group Using MicroCT. Group GC4/RECV GC8 GC-PTH8 BSL CNTL + 30.5% (p = 0.05) 8 WK CNTL 19.4% (p = 0.03) 25.6% (p = 0.00) GC4/RECV 27.8% (p = 0.01) Results are shown as percent change compared to groups on the left. Figure 4-16. Distal Femur Trabecular Separa tion by Group using MicroC T. a = significant compared to BSL CNTL; b = significant compared to 8 WK CNTL; c = significant compared to GC4/RECV. Results are reported as mean + standard error. Osteoblast Surface and Osteoclast Surface Osteoblasts and osteoclasts ar e the cells involved in reso rbing and forming new bone. They act in concert and the ba lance of numbers and activity le vels of these two cell types determines whether there is an overall increase or decrease in bone. Many disease conditions affect the numbers and activity levels of these ce lls. The ratio in number rand activity levels are also influenced by treatments, including PTH. Osteoblast and Osteoclast Surface data are presented in Table 4-10, Table 420, and Figures 4-17 and 4-18 There were no significant ageTrabecular Separation (Tb.Sp) 0 100 200 300 400 500 600 700 BSL CNTL 8WK CNTL GC4/ SAC GC4/ RECV GC8 GC4/GCPTH4 GC-PTH8 micrometers ab b ,c

PAGE 77

77 related changes in Ob.S or Oc.S in the distal femur as show n in Figure 4-17 and Figure 4-18. Ob.S tended to decrease with GC and increase with GC recovery or PTH. Trends in Oc.S were not as clear cut. Table 4-20. Significant Changes in Distal Femur Osteoblast Su rface and Osteoclast Surface by Group. Group GC4/SAC GC4/RECV GC8 GC4/GC-PTH4 GC-PTH8 BSL CNTL Oc.S + 116.7% (p = 0.02) Oc.S + 183.3% (p = 0.01) Ob.S + 153.2% (p = 0.01) 8 WK CNTL Oc.S 63.6% (p = 0.03) Ob.S 84.7% (P= 0.00) Ob.S -70.8 (p = 0.00) Ob.S + 177.8% (p =0.00) GC4/SAC Oc.S + 225% (p = 0.00) Ob.S + 690.9% (p = 0.00) Ob.S +1,136.4% (p= 0.00) Oc.S + 325.0% (p = 0. 00) Ob.S + 1,718.2% (p = 0.00) GC4/RECV Ob.S + 129.9% (p = 0.00) GC 8 Ob.S + 314.30 % (p = 0. 00) Ob.S + 547.6% (p = 0.00) Ob.S + 852.4% ( p= 0.00) GC4/GC-PTH4 Oc.S + 112.5% (p = 0.03) Ob.S + 47.1% (p = 0.05) Results are shown as percent change compared to groups on the left.

PAGE 78

78 Figure 4-17. Distal Femur Osteoblast Surf ace by Group using Histomorphometry. a = significant compared to BSL CNTL; b = si gnificant compared to 8 WK CNTL; c = significant compared to GC 4/SAC; d = significant compared to GC4/RECV; e = significant to GC8; f = significant compared to GC4/GC-PTH4. Results are reported as mean + standard error. Figure 4-18. Distal Femur Osteoclast Su rface by Group using Histomorphometry. a = significant compared to BSL CN TL; b = significant compared to 8 WK CNTL; c = significant compared to GC4/SAC; d = significant compared to GC4/GC-PTH4. Results are reported as mean + standard error. Osteoblast Surface (Ob.S) 0 5 10 15 20 25 30 BSL CNTL 8WK CNTL GC4/ SAC GC4/ RECV GC8 GC4/GCPTH4 GC-PTH8 % a, b, c d, e, f b, d c,e c, e Osteoclast Surface (Oc.S) 0 0.5 1 1.5 2 2.5 BSL CNTL 8WK CNTL GC4/ SAC GC4/ RECV GC8 GC4/GCPTH4 GC-PTH8 % a,c a,c,d b

PAGE 79

79 Dynamic Measures of Bone Fo rmation in the Distal Femur Dynamic measures included mineralizing surf ace (MS), mineral apposition rate (MAR) and surface referent bone formation rate (BFR/BS) Changes in bone formation were measured or calculated using unstained 8 m-thick sect ions. Data for MS, MAR, and BFR/BS are provided in Table 4-10. Differe nces in MS, MAR, and BFR/BS between groups are shown in Tables 4-22, Table 4-23 and Table 4-24 as well as in Figure 4-19, Figur e 4-20 and Figure 4-21, respectively. Mineralizing Surface Mineralizing surface was determined by measur ing the total perimete r of trabeculae and the percentage of the perimeter with double fluor ochrome labeling. Significant changes in MS are shown in Table 4-21. Mineralizing surface was lower in older anim als and those exposed to GC but tended to increase when GC was disconti nued and when PTH was used as shown in Figure 4-19. Animals in GC8 had 48.4% (p = 0.05) less MS than 8 WK CNTL, while GC4/SAC had 64.5% (p = 0.02) less MS than did 8 Wk CNTL. When GC treatment was discontin ued, MS increased. Compared to GC4/SAC, GC4/RECV had 436.4% (p = 0.0 0) higher MS and GC4/RECV had a 268.8%,p = 0.00) higher MS than did GC8. GC4/GC-PTH4 also showed a higher MS that 8 WK CNTL (+358.1%, p = 0.00), GC4/SAC (1,190.0%, (p = 0 .00), GC4/RECV (+140.7%, p = 0.01), and GC8 (787.5%, p = 0.00). Mineralizing surface wa s even more pronounced in the GC-PTH8 group where it was higher than 8 WK CNTL (441.9%), p = 0.00), GC4/SAC (1,427.3%, p = 0.00), GC4/RECV (184.7%, p = 0.0 0), and GC8 (950.0%, p = 0.00). MAR was lower in older animals but was not significantly decreased by GC use. However, MAR tended to increase when GC wa s discontinued and when PTH was administered

PAGE 80

80 Table 4-21. Significant Changes in Di stal Femur Mineralizing Surface by Group. Group GC4/SAC GC4/RECVGC8 GC4/GC-PTH4 GC-PTH8 BSL CNTL 83.8% (p = 0.01) 76.8% (p = 0.02) +111.9% (p = 0.01) + 150.7% (p = 0.00) 8 WK CNTL 65.4% (p = 0.02) + 90.3% (p = 0.05) 48.4% (p = 0.05) + 358.1% (p= 0.00) + 441.9% (p = 0.00) GC4/SAC + 436.4% (p = 0.00) + 1,190.0% (p = 0.00) +1,427.3% (p = 0.00) GC4/RECV + 140.7% (p = 0-.01) + 184.7% (p = 0.00) GC 8 +268.8% (P = 0.00) + 787.5% (p = 0.00) +950.0% (p = 0.00) Results are shown as percent change co mpared to groups on the left. Mineralizing Surface (MS)0 5 10 15 20 25 30 BSL CNTL8WK CNTL GC4/SACGC4/RECVGC8GC4/GCPTH4 GC-PTH8% b,c,e a,b, c, d, e a,b,c,d a,b a,b Figure 4-19. Distal Femur Mineralizing Surface by Group using Histomorphometry. a = significant compared to BSL CNTL; b = si gnificant compared to 8 WK CNTL; c = significant compared to GC 4/SAC; d = significant compared to GC4/RECV; e= significant compared to GC8. Re sults are reported as mean + standard error.

PAGE 81

81 Mineral Apposition Rate Significant changes in MAR are show n in Table 4-22 and Figure 4-20. Table 4-22. Significant Cha nges in Distal Femur Mineral Apposition Rate by Group. 8WK CNTL GC4/SAC GC4/RECVGC8 GC4/GCPTH4 GC-PTH8 BSL CNTL 33.3% (p = 0.01) -33.3% (p = 0.01) 33.3% (p = 0.02) 8 WK CNTL + 16.7% (p = 0.03) + 33.3% (p = 0.01) + 33.3% (p = 0.02) GC4/SAC + 16.7% (p = 0.03) + 33.3% (p = 0.01) + 33.3% (p = 0.02) GC4/RECV + 14.3% (p = 0.02) GC 8 + 33.3% (p = 0.02) Results are shown as percent change compared to groups on the left. Figure 4-20. Distal Femur Mineral Appositi on Rate by Group using Histomorphometry. a = significant compared to BSL CNTL; b = si gnificant compared to 8 WK CNTL; c = significant compared to GC 4/SAC; d = significant compared to GC4/RECV; e = significant compared to GC8. Results are reported as mean + standard error. Mineral Apposition Rate (MAR) -0.1 0.1 0.3 0.5 0.7 0.9 1.1 1.3 BSL CNTL 8WK CNTL GC4/ SAC GC4/ RECV GC8 GC4/GC-PTH4 GC-PTH8 m/d b,c a b,c,d,e b,c a,da,d

PAGE 82

82 as shown in Figure 4-20. Mineral Apposition Rate was 33.3% (p = 0.01) lower in 8 WK CNTL than in BSL CNTL. Mineral Apposition rate re mained the same betw een GC8 and 8 WK CNTL (0.6 m/day). However, in animals recovering from GC exposure, GC4/RECV had a 16.7% (p = 0.03) higher MAR than 8 WK CNTL and a 16.7% (p = 0.03) higher MAR compared with the GC4/SAC group. Exposure to PTH resulted in the GC-PTH gr oups having significantly higher MAR than control or GC-treated groups. Compared with comparably aged animals in the 8 WK CNTL group, MAR in the GC4/GC-PTH4 group was 33.3% (p = 0.01) greater. The MAR in GC/GCPTH4 group was also 33.3% (p = 0.01) highe r than in the GC4/SAC group, 14.3% (p = 0.02) higher than in GC4/RECV, and 33.3% (p = 0.02) hi gher than in GC8. MAR was greater in GCPTH8 than in 8 WK CNTL (+33.3%, p = 0.02) Bone Formation Rate/Bone Surface (BFR/BS) Both MS and MAR are measured directly, while the third dyna mic measure in bone, BFR/BS, was calculated. BFR/BS is derive d by multiplying MS x MAR and represents the average amount of bone formed daily (um3/um2/day). Significant changes in BFR/BS based on treatment groups are found in Table 4-23. Th ere was a non-significant trend toward lower BFR/BS in older animals and a significant decrease in BFR/BS following GC exposure in the GC4/SAC group compared with the 8 Wk CNTL group. However, there was no significant difference in BFR/BS between the GC8 and 8 WK CNTL group. PTH treatment did result in significantly higher BFR/BS. The GC4/PTH4 gr oup had a BFR/BS higher than the 8 WK CNTL (+390.9%, p = 0.01), the GC4/SAC (+1,700%, p = 0.00), the GC4/RECV (+157.1%, p = 0.02), and the GC8 (+980%, p = 0.00) groups. This pattern continued with the GC-PTH8 group.

PAGE 83

83 Table 4-23. Significant Change s in Distal Femur Bone Fo rmation Rate by Group using Histomorphometry. GC4/SAC GC4/RECV GC8 GC4/GC-PTH4 GC-PTH8 BSL CNTL 90.6% (p = 0.00) 84.4% (p = 0.01) + 131.3% (p = 0.02) 8 WK CNTL 72.7% (p = 0.02) + 90.9% (p = 0.03) + 390.9% (p = 0.01) + 572.7% ( p = 0.00) GC4/SAC + 600.0% (p = 0.00) + 1,700% (p = 0.00) + 2.366.7% (p = 0.00) GC4/RECV + 157.1% ( p = 0.02) + 254.4% (p = 0.00) GC 8 + 76.2% (p = 0.00) + 980.0% (p = 0.00) + 1,380.0% (p = 0.00) Results are shown as percent change co mpared to groups on the left. Figure 4-21. Distal Femur Bone Formation Rate/Bone Surface by Group using Histomorphometry. a = significant comp ared to BSL CNTL; b = significant compared to 8 WK CNTL; c = significan t compared to GC4/SAC; d = significant compared to GC4/RECV; e = significant co mpared to GC8. Results are reported as mean + standard error. Bone Formation Rate/Bone Surface (BFR/BS) 0 5 10 15 20 25 30 BSL CNTL 8WK CNTL GC4/ SAC GC4/ RECVGC8 GC4/ GCPTH4 GC-PTH8 um3/um2/d a,b,c, d,e a b,c,d,e a,bb,c,e

PAGE 84

84 Mid-Shaft Cortical Bone Data and Significant Changes A portion of the mid-shaft of the left femur in 70 mice was also analyzed using microCT. In each animal, 55 slices (1.2 m each) were scanne d and reconstructed. Cortical thickness (Ct.Th) data are presented in Table 4-24, Table 4-25, and Figure 4-22. Ther e was no apparent agerelated difference in cortical thickness. While Ct.Th did not change significantly with GC treatment, it did increase signifi cantly with PTH as shown in Fi gure 4-22. Animals receiving 8 weeks of GC and PTH had significantly hi gher Ct.Th than BSL CNTL (+8.7%, p = 0.02), GC8 (9.5%, p = 0.02), or GC4/SAC (+10.7%, p = 0.0 0). The only other significant difference occurred where Ct.Th was 6.7% higher (p = 0.0 3) in the GC4/GC-PTH4 group than in GC4/SAC animals. Table 4-24. Summary of Histomorphometri c Analysis of Femur Mid-Shaft by Group. BSL CNTL (n = 10) 8 WK CNTL (n= 10) GC4/ SAC (n = 10) GC4/ RECV (n = 10) GC 8 (n = 10) GC4/GCPTH4 (n = 10) GCPTH8 (n = 10) Cortical Thickness 27.5 + 1.9 28.3 + 2.4 27.0 + 1.4 28.2 + 1.3 27.3 + 1.7 28.8 + 1.7 29.9+ 2.0 Results are reported as mean + standard deviation. Table 4-25. Summary of Significant Changes in Femur MidShaft Cortical Bone Thickness by Group using MicroCT. GC4/GC-PTH4 GC-PTH8 BSL CNTL + 8.7% (p = 0.02) GC8 9.5% (p = 0.02) GC4/SAC4 + 6.7% (p = 0.03) 10.7% (p = 0.00) Note: Results are shown as percent change compared to groups on the left.

PAGE 85

85 Figure 4-22. Mid-Shaft Femur Cortical Thickn ess by Group using MicroCT. a = significant compared to BSL CNTL; b= significant w ith GC8; c = significant with GC4/SAC Results are reported as mean + standard error. Cortical Thickness (Ct.Th) 0 5 10 15 20 25 30 35 BSL CNTL 8WK CNTL GC4/ SAC GC4/ RECV GC8 GC4/GCPTH4 GC-PTH8 micrometers a,b,c c

PAGE 86

86 CHAPTER 5 DISCUSSION This was the first prospective study to evaluate the ability of PTH tr eatment to inhibit the negative skeletal effects of GC in mice. The study examined several aspects of this issue including the cumulative effects of both GC and PTH over time, site specific responses, tissue and cellular changes in response to treatment, and the effectiven ess of natural recovery from termination of GCs. The major findings of this investigation are as follows: GCs suppress bone formation/turnover but do not necessarily reduce bone volume PTH was bone anabolic even in the presence of GCs PTH increased bone mass quickly There was a residual effect following disconti nuation of GC therapy that resulted in no increase in bone mass despite a rebound e ffect in osteoblast numbers and activity Following GC treatment with PTH was more effective in improving bone structural parameters and mineralization than was natural recovery The magnitude of response to GC and PTH varied by skeletal site There was an age-related decline in bone st ructural parameters in control animals. Glucocorticoid Drugs Suppressed Bone Formation But Did Not Affect Bone Volume A previous study found that bone turnover de creased 67.4% (p = 0.05) in Swiss Webster mice treated with GC for 4 weeks (5). Other st udies have reported sim ilar findings (2,5,114) and it is generally accepted that exposure to GCs s uppresses bone turnover in mice. We also found that turnover decreased but, in contrast to so me studies, we observed increased BV/TV, though this only reached the level of statistical signifi cance in the distal femur and only in measurements using microCT. GCs may have caused an uncoupling of the re modeling process so there was no longer the same amount of bone formed as resorbed. In so me previous studies this resulted in a lower

PAGE 87

87 BV/TV in mice, but in the presen t study we detected a bone speci fic increase in BV/TV. While we found no significant changes in lumbar spine BV/TV after 8 weeks of GC use, we did detect (by microCT) a significant increase in BV/TV in the distal femur. At the cellular level, GCs affect osteoblasts, os teocytes, and osteoclast s but typically affect the osteoblast most (39). GCs typically decreas e osteoblast and osteocla st numbers and activity levels while also increasing oste ocyte apoptosis. The effects on os teoclasts are more variable. Studies invariably find Ob.S d eclines, but studies have found bot h increases and decreases in Oc.S (39). Lane found significant increases in Oc.S while Weinstein found increased and decreased Oc.S in different studies (2,5,114). In our study, 4 weeks of GC resulted in an 84.7% (p = 0.03) lower Ob.S and a 63.6% (p = 0.03) lowe r Oc.S in the distal femur in GC-treated animals. In the present study, decreased Ob.S had far-re aching consequences. Mineralizing surface decreased by over 60%, further s uggesting an important effect of GCs on osteoblast numbers. Mineralizing surface generally reflects the numbe r of osteoblasts on trabecular surfaces and histomorphometry detected a trend toward decreas ed Ob.S and Oc.S in the GC8 group compared to 8 WK CNTL although the changes did not reach st atistical significan ce. It is possible that the increased BV/TV seen in this study results from GCs having more of a suppressive effect on osteoclasts than on osteoblasts. If osteoclast activity levels were suppr essed more, relative to osteoblast levels, this would resu lt in increased bone volume. GCs not only affect bone cell numbers, they also influence the activity levels of these cells. In osteoblasts, GCs reduce the tim e these cells secrete osteoid a nd actively mineralize bone (72). Although we did not measure osteoid, measurement of dynamic parameters suggest a decrease in bone turnover. In the distal femur, we f ound MS was 48.4% (p = 0.05) lower in GC-treated

PAGE 88

88 animals (GC8) than in 8 WK CNTL animals but found no difference in MAR, an index of osteoblast activity, and only a non-si gnificant, two-fold decrease in BFR/BS. These findings are similar to other studies with respect to MS but differ with studies that found lower MAR and BFR/BS (2,5). Our findings with respect to MAR and BFR/BS are consistent with the differences in BV/TV we detected, however. It is more difficult to directly assess osteoclast activity levels, but inferences can be made. The typical response to GCs in human disease is decreased bone turnover that disproportionately affects formation rather than resorption. This resu lts in rapid bone loss. Most studies of GC in mice have found that BV/TV declines in a dose-dependent and time-dependent manner. This was the case in several studies, including some using Swiss Webster mi ce A previous study found that 6-month old Swiss Webster mice treated for three weeks had a modest 19% (p = 0.05) lower BV/TV in the lumbar vertebrae based on histomorphometry and 22% (p = 0.05) decrease based on microCT (2). In this same study, Oc.S increased over 100 %, from 0.8 to 1.7% (p < 0.05), MS decreased by almost one-t hird, from 42.9 to 29.5% (p < 0.05) while MAR declined almost 40% from 1.03 to 0.64 m (p < 0.05). These changes accompanied a 105% increase in osteocyte apoptosis from 1.9 to 3.9% (p < 0.05) (2). Findings such as these have previously been attributed to a decrease in osteoblastogenesis and increased apoptosis of both osteoblasts and osteocytes (2,5,39). What determines the net balance of formation to resorption is unclear but our finding of increased BV/TV in mice should not be dism issed without further study since the same phenomenon has been seen in other rodents. In rats, researchers have detected bone loss, bone increase, or no change after expos ure to GCs (36-38). There has al so been variability reported in mice exposed to GCs. While some studies in mice have reported doseand time-dependent

PAGE 89

89 losses in BV related to GC use, others ha ve not. Using histomor phometry, one study found no significant decline in BV/TV even though BFR/BS declined signifi cantly (~75%) in the femur. The only significant differences detected by micr oCT occurred at very high GC dosages, which are known to impact other steroid receptors (3). This suggests that the dose-response to GCs is not linear and that some conditions may cause reso rption to be more suppressed than formation. Our findings suggest, that while changes in bone cell numbers were important, changes in activity levels at the cell ular level may also be central to the changes we detected in these bones. While both osteoblasts and osteoc lasts decreased in number in re sponse to GC treatment, the increased BV/TV found with GC treatment found in our study suggests oste oclast activity levels may have been more severely suppressed relative to osteoblast activity, allowing for increased BV/TV despite decreased numbers of osteoblasts. This requires further examination, however, since there were no measures in this study to detect or quantify oste oclast activity levels. Anabolic Effects of PTH Prevented the Inhibi tory Changes Associated with Glucocorticoid Drugs Studies evaluating the effects of PTH on mice co nsistently show it is bone anabolic even under conditions that tend to decrease bone fo rmation and/or bone ma ss (7,8,45) and our study supports this finding as well. In one study involving ovariectomized mice, PTH treatment resulted in a two-fold increase in MAR a nd a 3-fold increase in BFR/BS (7). GCs suppress osteoblast prolifera tion/ activity levels and PTH, a hormone that stimulates osteoblasts, is able to reverse th is trend (6) to overcome the s uppressive effects of GCs. PTH and GC together have been tested in rats (115), but the present study is the first to examine the effects of PTH in GC-treated mice. The only huma n investigation of the combined effects of GC and PTH involved subjects who also took estr ogen. In that study, BMD increased linearly during the entire 12 month course of treatment (6). This finding was consistent with other

PAGE 90

90 studies that found that the antire sorptive effects of estrogen di d not prevent the bone anabolic response to PTH (116). The anabolic nature of intermittent PTH on bone is seen in its effects at the cellular level. Following PTH administration to rats and humans, th ere is typically an increase in both Ob.S and Oc.S, leading to increased bone turnover. Althou gh both Oc.S and Ob.S increase, this change is most dramatic in Ob.S and, therefore, favors bone formation. A study of PTH in intact mice reported a significant increase in Ob.S (45). We found no significant change in Oc.S but a significant increase in Ob.S compared to th e GC8 group after 4 weeks (+ 547.6%, p = 0.00) and 8 weeks (+ 852.4%, p = 0.00) of PTH treatment. This substantial change in osteoblast presence on trabecular surfaces could reflect increased di fferentiation, decreased a poptosis, or reactivation of bone lining cells. Consistent with the increas e in Ob.S, after 4 weeks of treatment with PTH, there was a significant increas e in MS (787.5%, p = 0.00) MAR (33.3%, p = 0.02), and BFR/BS (980.0%, p = 0.00) compared to animals receiving ei ght weeks of GC. The differences in MS and BFR/BS were even more pronoun ced after 8 weeks of PTH. In this study, Ob.S was over 1,000% higher in a group of animals treated with PTH for 4 weeks compared to animals trea ted similarly with GCs (GC4/SAC ) The increase in number and activity levels of the osteoblasts resulted in a BFR/BS over ten times the rate seen in GC8 animals in just 4 weeks. Likewise, animals treated with PTH and GC for 8 weeks had over 850% greater Ob.S than did GC8 animals. The magnitude of change in Ob.S makes it unlikely that it could result solely fr om reactivation of quiescent bone lining cells. The very high increases in Ob.S seen with PTH likely stem fr om increased osteoblast proliferation as well as increases in osteoblast lifespan (116).

PAGE 91

91 Mineralizing surface is an i ndirect measure of the osteob last population and increased numbers of these cells means more bone su rface is mineralizing at any given time. Mineralization apposition rate ge nerally reflects the activity le vel of the osteoblasts, since osteoblasts modulate bone mineralization. The si gnificantly higher MS and MAR detected in the present study are evidence PTH increased both th e number and activity levels of osteoblasts. BFR/BS, a function of both MS and MAR, is al so significantly higher wi th PTH, reflecting the shifting balance in bone turnover toward bone formation. There are only a few studies that have examined the effects of PTH following GC th erapy on BFR/BS. One of these found that BFR/BS tended to decrease in response to GC al one but increased with PTH alone in rats. That study found that when the two drugs are used simu ltaneously, there was an intermediate response that increases BFR/BS relative to GC-only treated animals, but kept it lower than with PTH alone (115). The effects of PTH also appeared to increa se over time. Animals in the GC-PTH8 group showed a trend toward even higher bone mass than the GC4/GC-PTH4 animals, according to microCT, but high variability prevented this from reaching statistical signi ficance. MicroCT found a 45.6% (p = 0.02) greater BV/TV in distal femur between animals treated with PTH for 8 weeks versus those treated for only 4 weeks. These changes appear driven by additional increases in osteoblast and osteoclast numbers since Ob.S was 47.1% (p = 0.05) higher and Oc.S was 112.5% (p = 0.03) higher in group receiving 4 a dditional weeks of PTH. At the end of the 8 weeks of PTH treatment, osteoclasts lined 1.7% of the trabecular surfaces while osteoblasts lined 20% of the trabecular surfaces. PTH Increases Bone Mass Quickly Studies have found the effects of PTH are ra pid and one study found si gnificant changes in BMD in the tibia of C57BL/J6 mice with just one week of treatment (45), although it took longer

PAGE 92

92 to detect BMD differences in the vertebrae. In the present study, there was also significantly higher Ob.S, MS, MAR, and BFR/BS in the di stal femur after 4 weeks of PTH-treatment indicating increases in osteoblast numbers and activity leve ls quickly influenced both bone formation and mineralization activity. These impr ovements were evident in increased vertebral BV/TV and Tb.Wi at 4 weeks based on histomorpho metry and in Tb.Th based on microCT. Our study mirrors other studies in finding early and significant changes in dynamic indices of bone turnover but a lag of at leas t 4 or more weeks before structur al differences re ach statistical significance difference from baseline controls (8,45). Changes in bone structural parameters were brought about through the ability of PTH to increase bone turnover even though GC was also administered. Animals receiving PTH for 4 weeks showed increases in Ob.S (+547.6%, p = 0.00) but a non-significa nt decrease in Oc.S (-11.1%, p = ns) compared to GC8 animals. This helps show why PTH is anabolic since the proportion of the trabecular surfaces covered in osteoblasts increased over 5-fold. PTH also increased the activity levels of osteoblasts as reflected in higher MS, MAR, and BFR/BS. Of interest, animals treated with PTH for 8 weeks while also receiving GCs, showed a further increase in Ob.S (+47.1%, p = 0.05) and Oc.S (+112.5%, p = 0.03) compared to animals in the GC4/GC-PTH4. Residual Effects of Glucocor ticoid Drugs are Apparent During Natural Recovery No previous studies have eval uated bone recovery once GC trea tment is discontinued. We expected discontinuation of GC tr eatment would result in natural recovery that would bring the measured parameters closer to control animals. That was not the case since we detected trends toward lower BV/TV, Tb.N and Tb.Th and high er Tb.Sp in the GC4/RECV group than in the GC4/SAC, a group of animals sacrificed immediat ely following four weeks of GC treatment. This suggests there is a residual e ffect to GC use that, at least in mice, does not abate in four

PAGE 93

93 weeks. Although there were no st atistically significant changes in bone structural parameters, there were significant changes in activity at the cellu lar level. Four weeks of recovery resulted in significantly higher Oc.S (+225%, p = 0.00) a nd Ob.S (+ 690.9%, p = 0.00) compared to GC4/SAC. There was a rebound effect in dynami c bone measures, that resulted in MS, MAR, and BFR/BS exceeding even the 8 WK CNTL gr oup. Increases in these parameters, however, did not translate into increased bone mass, suggesting either there was not sufficient time for increased bone mass to be detected or ther e were GC-induced changes to the bone matrix preventing expected increases. We believe the latter is more likely and th at this may represent a change to the bone matrix that inhibits increases in structural bone mass even once GC treatment is discontinued. The reason for the impaired ability of these bones to increase in mass despite high activity levels in bone cells is unclear, but is not without precedent. Experiments in animals treated with alcohol show similar impaired recovery. When decalcified bone cores from alcohol-treated and control mice were placed subcut aneously in untreated mice, th e alcohol-exposed cores showed an impaired ability to re-mineralize (117). It is possible a similar process is at work in glucocorticoid-treated mice. Conve rsely, it is also possible a l onger recovery period would have made a difference. After GCs are discontinued in humans there is a sl ow reduction in fracture risk over time (23). One study found the relative fracture risk decrease d from 2.4 to 1.8 one year after GC is stopped, and continued to move back to pre-treatment baseline risk levels over time (23). These reductions in fractur e risk may result from improved structural repairs to the bone over time. PTH Treatment After Glucocort icoid Use Was More Effective than Natural Recovery We expected administration of PTH following GC treatment to be more effective in increasing bone structural parameters than natu ral recovery, and the data support this. In the

PAGE 94

94 vertebrae, Tb.Th was increased by 25.5% (p = 0.00) using histomorpho metry and by 13.5% (p = 0.00) using microCT in GC4/GC-PTH4 compared w ith GC4/SAC. In the distal femur, BV/TV and Tb.Wi were significantly higher in GC4/GC -PTH4 compared with GC4/SAC while Tb.Sp was significantly lower based on histomorphometry. In the di stal femur, BV/TV increased significantly (95.6% (p = 0.01) although there was only a non-significant increase in Ob.S between the GC4/GC-PTH4 and GC4/SAC groups Histomorphometry did, however, detect significant increases in MS (140.7%, p = 0.01), an indication that a greate r percentage of the trabeculae were covered in osteoblasts. There was also a 14.3% (p = 0.02) increase in MAR, a measure of osteoblast activity levels. The in creases in MS and MAR contributed to the significant increase in BFR/BS (+ 157.1, p = 0.02) indicating that PTH treatment resulted in increased bone formation. These findings were expected; since PTH in bone anabolic we expected it would have a greater effect on bone parameters than natural recovery after GCtreatment was discontinue d. While animals that experienced natural recovery showed a rebound effect in dynamic bone parameters that show ed significant increases in MS(90.3%, p = 0.05), MAR (16.7%, p = 0.03), and BFR/BS (90.9%, p = 0.03) compared to 8 WK CNTL, these levels were much more impressive following PTH treatment: MS (441.9%, p = 0.00), MAR (33.3%, p = 0.02), BFR/BS (572.7%, p = 0.00). Age-Related Effects on Bone Mass We expected a relatively constant level of bone turnover in these adult male Swiss Webster mice. A previous study, whose goal was specifi cally to determine when long-bone growth ceased, found that this strain of mice cease l ongitudinal bone growth in the long bones between five and six months of age (5). Based on this we expected young adult Swiss Webster mice to have a constant level of bone turnover that woul d result in stable levels of bone mass throughout the two-month length of the study. Our data sugges t, however, this was not the case. Our results

PAGE 95

95 may represent the first reporting that age-relate d decreases in bone mass occur in Swiss Webster mice begin within a few months of these animals reaching peak bone mass. The effect was seen in histomorphometry of the lumbar spine and bot h histomorphometry and mi croCT of the distal femur. There is evidence of variability between and w ithin in-bred mouse strains with respect to skeletal maturation (118,119) and, this study sugg ests there may be perh aps also age-related changes to bone in young adult out-bred mice, like the Swiss Webster st rain. One study found that the Swiss Webster mouse reaches skeletal maturity between 5 and 6 months of age (5), although no attempt was made to determine the age at which there would be a natural decline in bone mass due to aging. Another study found that the level of osteobl astogenesis decreased three-fold in SAMP6 mice between 3 and 4 months of age and resulted in significant bone loss soon after (120). Data from another study usi ng 10 week old C57BL/J6 mice showed a trend toward decreasing bone volume at about 16 weeks of age (45). Our data also suggest an agerelated decrease in bone mass in control animals. The Swiss Webster mice in our study entered at seven months of age and were sacrificed tw o months later. There was a lower cancellous BV/TV in both the lumbar spine and distal femur, using microCT and histomorphometry, between 7 month old and 9 month old control animals. In the distal femur, both histomorphomet ry and microCT found 35-45% less BV/TV and around a 20-40% lower Tb.N in nine month ol d animals than in their seven month old counterparts. There was also more than doubl e the Tb.Sp according to histomorphometry. Despite these changes there was no significant di fference in the Ob.S and Oc.S although Ob.S showed a downward trend while Oc.S showed an upward trend. The dow nward trend in Ob.S was insufficient to cause a significant differe nce in MS. However, there was a significant

PAGE 96

96 decline in MAR, suggesting that, while osteobl ast numbers may be re latively stable, the osteoblasts were less actively forming bone. Th e 33.3% (p = 0.01) declin e in MAR contributed to the almost three-fold decline in BFR/BS in the 8 WK CNTL compared to the BSL CNTL animals. Prophylactic Value of Concurrent Treatme nt with Glucocorticoid Drugs and PTH This study is the first to examine whether there is increased be nefit to starting PTH treatment concurrently with GC administration as opposed to using GC first and then later treating with PTH. One group of animals was treated with GC for 4 weeks and then treated for an additional four weeks with both GC and PTH. Another group simultaneously received GC and PTH for 8 weeks. There was a benefit to si multaneous treatment but this may be the result of the additional 4 weeks of PTH. There was a 47.1% (p = 0.05) higher Ob.S and a 112.5% (p = 0.03) higher Oc.S with 8 weeks of PTH, indicati ng continued gains in the latter group. In the lumbar vertebrae, the group treated with PTH for 8 weeks also had a significant decrease in Tb.Sp. In the distal femur, significant change s were only seen with microCT, where both BV and Tb.Th were significantly higher. There were no significant differen ces in MS, MAR, or BFR/BS with additional PTH treatment. These findings suggest that even though 8 weeks of PTH resulted in some additional improvement to bone parameters, 4 weeks of PTH after GC exposure was still highly effective. Comparing the effects of 4 weeks of PTH us e against eight weeks of PTH in animals receiving GC for 8 weeks shows that most of th e change occurred in the first 4 weeks of PTH administration. There were continued improveme nts in the group receiving PTH for 8 weeks, but the magnitude of change was less. 82.9% of the final level of MS seen in the GC-PTH8 group was seen in the GC4/GC-PTH4 group. Th e same trend was true for MAR which was slightly higher in the 4 week PTH group and BFR/BS where over 70% of the total 8 week

PAGE 97

97 change was seen in the GC4/GC-PTH4 group. Th ese findings suggest PTH was able to reverse the effects of prior GC use sufficiently to make it a valid post-exposure treatment. Site Specificity of GC and PTH Treatment In mice, as in humans, BV/TV is higher in th e vertebrae than in the femur. In humans however, the effects of both GC a nd PTH are more prevalent in the vertebrae than in the femur. Humans treated with both GC and PTH increased spinal BMD by 35% while the hip gained only a modest 2% BMD over a one-year period (6). A different relati onship has been reported in mice (8) where there is greater change in long bones of the a ppendicular skeleton than in the axial skeleton. Other studies in rats have reported the same results (39). The difference in results found in rodents may result from biomechan ical loading differences between bipeds and quadrupeds (8). In our study, changes with both GC and PTH were also greater in the distal femur than in the lumbar spine. Based on histomorphometry, BV/TV changes were greater in the distal femur than in the lumbar spine (-46% versus -30%) and based on PTH exposure (100% versus 42%). It may be that the biomechanics of weight-bearing activitie s of the mouse explain the different findings or that th ere is an additive effect when PTH is acting on a bone already exposed to higher mechanical loading (8,121,122). Conclusions This study found that glucocorticoid drugs suppress bone turnover in the Swiss Webster mouse and that these effects can be offset by PTH. The effects of PTH were rapid, with most of the total changes seen ov er eight weeks occurring in the first 4 weeks. We also found that PTH treatment after GC exposure is more effective at restoring bone mass than natural recovery. We found there are site specif ic differences in response to PTH and that these caused greater changes to occur in the appendicular rather than in the axial skeleton.

PAGE 98

98 This study also reports some novel findings. We found there was an age-related loss of bone between 7 and 9-month old animals who re ceived only vehicle and no study drugs. We also found that GCs tended to increase bone ma ss even though there was clearly a suppression of bone turnover. Finally we found a residual effect with GC treatment that inhibited increases in bone mass after GC treatment was discontinue d even though osteoblastic activity rebounded after GC removal. Clinical Applications The underlying goal for this research was to demonstrate the efficacy of using PTH to mitigate the skeletal effects of GCs in mice, as a prelude to studies in humans. Swiss Webster mice have previously been validated as models of bone loss due to ovariectomy (7) and for GCinduced bone loss (2) and have b een described as a faithful mode l of the glucocorticoid-induced bone loss in humans(39). Despite the fact this study found exposur e to GC may have increased BV/TV, we believe the results pave the way fo r follow-on studies using PTH to treat bone loss due to GCs in humans, although our results suggest the response of Swiss Webster mice to GCs may be more variable than previously reported. There has been ample research demonstrating the deleterious effects of GCs and the increased risk of bone fracture that accompanies their use (21,23,73,76,81). This study demonstrated that PTH is effective in reversing the inhibitory effects of GCs on bone formation. Our findings support a clinical a pplication for PTH in patients diseases conditions such as rheumatoid ar thritis, COPD and other conditions that result in long-standing GC use. PTH was used in one st udy with rheumatoid arthritis patients where it increased spinal BMD by 35%, although gains in th e hip were a more modest 2% (6). This human study concluded that while anti-resorptive treatments can prevent bone loss, PTH was the only therapy currently available that could reve rse the suppressive effects of GCs (6). We believe these findings and our current study show there is potential for successful treatment of

PAGE 99

99 other patients who have experienced long-term gl ucocorticoid therapy including those that may require solid organ transplants. Due to the nature of post-transplant immunosuppressant protocols and reports of osteosarcoma in rats exposed to life-time dos es of PTH (103-105), we would not recommend the use of PTH after transplant surgery, but this dr ug may prove efficacious in treating pre-existing, GC-induced osteoporosis in those still awaiting su rgery. The rapid effects of the PTH suggest that even if patients could not complete th e normal 18-24 month treatment regimen prior to receiving a transplant, even a few months of tr eatment could be benefici al. Although beneficial effects are more likely to be seen in the spin e, one study showed that bone mass in the hip increased six months after PTH was discontinued (44), indicating therapeutic benefits beyond the dosing period. Study Limitations As with any study, resource considerations influenced study design and implementation. Animals were housed one to a cag e although this was not ideal. Mice are social animals that interact with each other so housing them alone might have affected activity or stress levels. We had to house the mice singly, however, to avoid fighting. These were adult males formerly used for breeding and were highly territorial. We also had to accept delivery of study animals in 7 shipments of 10 mice rather than receiving a ll animals simultaneously. This was necessary because of the age of the animals involved. Th e other alternative would have been to buy young animals and age them either at our facility or with the vendor. This would have decreased cohort variation but was cost-prohibitive. We minimi zed the impact of multiple cohorts by assigning animals from each arriving shipment to each study group.

PAGE 100

100 Future Directions There were novel findings in this study that wa rrant further investigation. The finding of negative changes in bone structural parameters in control mice over the two-month course of the study was unexpected. Further studies are needed to verify these differences and establish the significance and mechanisms mediat ing these changes. If there is a natural loss of bone so quickly after attaining peak bone mass, researcher s need to be aware of this so they dont attribute age-related changes to interventional treatments. The residual effect of GCs on the bone matrix also needs further study. Animals treated with GCs and then allowed to recover showed fu rther declines in bone structural parameters despite increased osteoblastic activity. It is possible bone structural parameters would have returned to normal if more recovery time had b een allowed, but it is also possible that the GCs caused a change to the bone matrix that inhibited recovery. If the latter is true, it might partially explain the increased risk of bone fracture after GC therapy. Finally, we believe the results of this study provide clear ev idence of the efficacy of PTH in treating bone changes caused by GC exposure. We believe human clinical populations may also benefit from this treatment as previously discussed.

PAGE 101

101 APPENDIX A SUMMARY OF BONE MEASUREMENTS A-1. Summary of Significant Changes in Lumb ar Vertebrae L3 based on Histomorphometry. BSL CNTL 8WK CNTL GC4/SAC GC4/ RECV GC8 GC4/GC-PTH4 GC-PTH8 BSL CNTL BV/TV -30.7% (p = 0.03) Tb.Wi -18.4% (p = 0.02) BV/TV -27.7% (p = 0.03) Tb.Wi -19.0 (p=0.02) BV/TV -31.5 (p = 0.02) Tb.Wi -25.1 (p = 0.00) 8 WK CNTL Tb.Wi + 24.5 % (p =0.01) BV/TV + 42.0% (p = 0.01) Tb.Wi + 22.1 (p = 0.01) Tb.Sp -24.7 % (p = 0.03) GC4/ SAC Tb.Wi + 14.5% (p = 0.03) GC4/RECV Tb.Wi +25.5% (p = 0.00) BV/TV +35.9% (p = 0.03) Tb.Wi +23.1% (p = 0.01) GC 8 Tb.Wi + 18.5 (p = 0.02) BV/TV + 26.4% (p = 0.03) Tb.Wi + 35.8% (p = 0.00) BV/TV + 43.7% (p = 0.01) Tb.Wi + 33.2% ( p = 0.00) GC4/GC-PTH4 Tb.N + 19.5% (p =0.02) Tb.Sp 17.4% (p = 0.01)

PAGE 102

102 A-2. Summary of Significant Changes in Lumbar Vertebrae L2 based on MicroCT. GC4/RECV GC4/GC-PTH4 GC-PTH8 BSL CNTL BV/TV 21.1% (p = 0.03) Tb.N 9.5% (p = 0.04) Tb.Th 5.7% (p = 0.02) Tb.Sp + 12.7% (p = 0.03) Tb.N -9.5% (p = 0.02) Tb.Sp + 12.3% ( p = 0.04) Tb.Th + 12.3% (p = 0.01) 8 WK CNTL Tb.Th + 9.7% (p = 0.04) BV/TV + 36.6% (p = 0.01) Tb.Th + 15.2% (p = 0.00) GC4/SAC Tb.Th + 10.6% (p = 0.01) BV/TV + 26.4% (p = 0.04) Tb.Th + 16.1% (p = 0.00) GC4/RECV Tb.Th + 13.5% (p = 0.00) BV/TV + 46.7% (p = 0.00) Tb.Th + 19.2% (p = 0.00) GC 8 BV/TV 21.1% (p = 0.04) Tb.Th + 12.3% (p = 0.00) Tb.Th + 17.9% (p = 0.00)

PAGE 103

103 A-3. Summary of Significant Changes in th e Distal Femur based on Histomorphometry. 8WK CNTL GC4/RECV GC8 GC4/GC-PTH4 GC-PTH8 BSL CNTL BV/TV 45.8% (p = 0.03) Tb.N 40.0% (p = 0.05) Tb.Sp + 153.0% (p = 0.05) BV/TV 52.1% (p = 0.01) Tb.N 40.0% (p = 0.01) Tb.Wi 22.3% (p = 0.04) Tb.Sp + 177.2% (p = 0.01) Tb.Wi 17.7% (p = 0.05) 8 WK CNTL BV/TV + 100.0% (p = 0.02) Tb.N + 83.3% (p = 0.03) Tb.Sp -59.4% p = 0.03 GC4/RECV BV/TV + 95.6% (p = 0.01) Tb.Wi + 27.4% (P =0.05) Tb.Sp 63.3% (p = 0.01) BV/TV + 126.1% (p = 0.01) Tb.N + 83.3% (p = 0.03) Tb.Wi + 30.6% (p = 0.02) Tb.Sp 63.0% (p = 0.03) GC 8 Tb.Wi + 20.3% (p =0.04) Tb.Wi + 23.3% P = 0.03

PAGE 104

104 A-4. Summary of Significant Changes in the Distal Femur based on MicroCT 8WK CNTL GC4/RECV GC8 GC4/GC-PTH4 GC-PTH8 BSL CNTL BV/TV 37.4% (p = 0.00) Tb.N 22.2% ( p=0.05) BV/TV 39.4% (p = 0.00) Tb.Sp +30.5% (p = 0.05) Tb.Th + 16.3% (p = 0.01) 8 WK CNTL BV/TV + 40.3% ( p = 0.03) Tb.N + 33.3% (p = 0.03) Tb.Th 12.7% (p = 0.04) Tb.Sp 19.4% (p = 0.03) BV/TV + 85.5% (p = 0.00) Tb.N + 33.3% (p = 0.00) Tb.Sp 25.6% (p = 0.00) GC4/SAC Tb.Th + 6.3% ( p = 0.05) Tb.Th + 17.5% (p = 0.00) GC4/RECV BV/TV + 45.0% ( p =0.02) BV/TV + 91.7% (p = 0.00) Tb.N + 27.3% (p =0.01) Tb.Th + 17.9% (p = 0.00) Tb.Sp 27.8% (p = 0.01) GC 8 Tb.Th + 13.5% p = 0.05 BV/TV + 32.3% p = 0.04 Tb.Th +25.4% P = 0.00 GC4/ GC-PTH4 BV/TV + 45.6% ( p = 0.02) Tb.Th + 10.5% (p = 0.05)

PAGE 105

105 A-5. Percent Changes in Osteoclast and Os teoblast Surfaces in the Distal Femur. p GC4/SAC GC4/RECV GC8 GC4/GC-PTH4 GC-PTH8 BSL CNTL Oc.S + 116.7% (p = 0.02) Oc.S + 183.3% (p = 0.01) Ob.S + 153.2% (p = 0.01) 8 WK CNTL Oc.S 63.6% (p = 0.03) Ob.S 84.7% (P= 0.00) Ob.S -70.8 (p = 0.00) Ob.S + 177.8% (p =0.00) GC4/SAC Oc.S + 225% (p = 0.00) Ob.S + 690.9% (p = 0.00) Ob.S +1,136.4% (p= 0.00) Oc.S + 325.0% (p = 0. 00) Ob.S + 1,718.2% (p = 0.00) GC4/RECV Ob.S + 129.9% (p = 0.00) GC 8 Ob.S +314.3% (p = 0. 00) Ob.S + 547.6 (p = 0.00) Ob.S + 852.4% ( p= 0.00) GC4/GC-PTH4 Oc.S + 112.5% (p = 0.03) Ob.S +47.1% (p = 0.05)

PAGE 106

106 A-6. Percent Changes in Dynamic Bone Form ation Parameters in the Distal Femur. 8WK CNTL GC4/SAC GC4/RECV GC8 GC4/GC-PTH4 GC-PTH8 BSL CNTL MAR 33.3% (p = 0.01) MS 83.8% (p = 0.01) MAR 33.3% (p = 0.01) MS 76.8% (p = 0.02) MAR 33.3% (p = 0.02) BFR/BS 84.4% (p = 0.01) MS +111.9% (p = 0.01) MS + 150.7% (p = 0.00) BFR/BS + 131.3% (p = 0.02) 8 WK CNTL MS 64.5% (p = 0.02 BFR/BS 72.7% (p = 0.02)) MS + 90.3% (p = 0.05) MAR + 16.7% (p = 0.03) BFR/BS + 90.9% (p = 0.03) MS 48.4% (p = 0.05) MS + 358.1% ( p= 0.00) MAR + 33.3% (p = 0.01) BFR/BS + 390.9% (p = 0.01) MS + 441.9% (p = 0.00) MAR + 33.3% (p = 0.02) BFR/BS + 572.7% ( p = 0.00) GC4/SAC MS + 436.4% (p = 0.00) MAR + 16.7% (p = 0.03) BFR/BS + 600.0% (p = 0.00) MS + 1,190.9% (p = 0.00) MAR +33.3% (p = 0.01) BFR/BS + 1,700.0% (p = 0.00) MS + 1,427.3% (p = 0.00) MAR + 33.3% (p = 0.02) BFR/BS + 2,366.7% (p = 0.00) GC4/RECV MS + 140.7% (p = 0-.01) MAR + 14.3% (p = 0.02) BFR/BS + 157.1% (p = 0.02) MS + 184.7% (p = 0.00) BFR/BS + 254.4% (p = 0.00) GC 8 MS +268.8% (p = 0.00) BFR/BS +76.2% (p = 0.00) MS +787.5% (p = 0.00) MAR + 33.3% (p = 0.02) BFR/BS + 980.0% (p = 0.00) MS + 950.0% (p = 0.00) BFR/BS + 1380.0% (p = 0.00)

PAGE 107

107 APPENDIX B SUMMARY OF SELECTED STUDIES IN MICE

PAGE 108

108 Table B-1. Studies Of Glucocorticoid-Induced Bone Loss In Mice. Strain Treatment Dose (mg/kg) Age (wk.) Gender #/Grp Time (days) Results Type Analysis SW Prednisolone (pellet) 2.1 28 M 4-5 27 Spinal BMD decrease Preferential loss in axial skeleton Increased resorption Decreased formation Increased osteocyte apoptosis Weinstein (5) 1998 DXA Histomorphometry SW Prednisolone (pellet) Prednisolone + alendronate 2.1 16 M 5-9 4, 10, and 27 Decreased Osteoclast apoptosis Increased osteoclast survival Decreased bone formation rate Increased osteoblast apoptosis Weinstein (4) 2002 DXA Histomorphometry

PAGE 109

109Table B-1. Continued. Strain Treatment Dose (mg/kg) Age (wk.) Gender #/grp Time (days) Results Type Analysis Balb/C Dexa-methasone (IP) 1 and 10 28 F 5 21 Changes only seen at higher dosages -Decreased BFR/BS and MAR Decreased osteocalcin McLaughlin (3) 2002 Histomorphometry MicroCT Biochemical Assays SW Prednisolone (pellets) 1.4 24 M Unk 21 -Decrease Tr bone volume and strength -Increased size of osteocyte lacunae -Increased DPD crosslinks (resorption) -Decreased Osteocalcin (formation) Lane (2) 2006 Histomorphometry MicroCT Biochemical Assays BFR/BS = bone formation rate/bone surface; BMD = bone mineral de nsity; DPD = deoxipyrodinoline; GC = glucocorticoi d; Tr = trabe cular; MAR = mineral apposition rate; SW = Swiss Webster.

PAGE 110

110Table B-2. Studies using Teriparatide in Mice. Strain Condition Dose (g/kg) Age (wk.) Gender N/ grp Time (Wks) Results Type Analysis C57BL/6 Ovx/sham 40 12 F 4-6 3 or 7 Increased bone formation rate and BV/TV Increased Oc. S -greater effect in LV than tibia Little change in cortical bone Zhou (8) Histomorphometry C57BL/6 Intact 40 10 F 9 3 and 7 -greater BMD change in tibia and femur than LV increased bone turnover Iida-Klein (45) Piximus (DXA) Histomorphometry SW Ovx 80 11 F 9 4 -Bone loss reversed Mice lost bone faster than rats 2-3X increase MAR Alexander (7) Histomorphometry MicroCT BMD = bone mineral density; BV/TV = bone volume/total volume; LV = lumbar vertebrae; MAR = mineral apposition rate; Oc. S = ost eoclast surface; Ovx = ovariectomized; SW = Swiss Webster.

PAGE 111

111 LIST OF REFERENCES 1. Hofbauer LC, Gori F, Riggs BL, Lacey DL, Dunstan CR, Spelsberg TC, Khosla S 1999 Stimulation of osteoprotegerin ligand and inhibition of osteopro tegerin production by glucocorticoids in human osteoblastic lineage cells: potential paracrine mechanisms of glucocorticoid-induced os teoporosis. Endocrinology 140 (10) : 4382-9. 2. lane NE, Yao W, Balooch M, Nalla R, Bal ooch G, Habelitz S, Kinney J, Bonewald LF 2006 Glucocorticoid-treated mice have locali zed changes in trabecu lar bone material properties and osteocyte lacuna r size that are not observed in placebo-treated or estrogendeficient mice. Journal of Bone and Mineral Research 21 (14) : 466-476. 3. McLaughlin F, Mackintosh J, Hayes BP, McLaren A, Uings IJ, Salmon P, Humphreys J, Meldrum E, Farrow SN 2002 Glucocorticoid-i nduced osteopenia in the mouse as assessed by histomorphometry, microcomputed tomography, and biochemical markers. Bone 30 (6) : 924-30. 4. Weinstein RS, Chen JR, Powers CC, Stewar t SA, Landes RD, Bellido T, Jilka RL, Parfitt AM, Manolagas SC 2002 Promotion of oste oclast survival and antagonism of bisphosphonate-induced osteoclast apoptos is by glucocorticoids. J Clin Invest 109 (8) : 1041-8. 5. Weinstein RS, Jilka RL, Parfitt AM, Manolag as SC 1998 Inhibition of osteoblastogenesis and promotion of apoptosis of osteoblasts and osteocytes by glucocorticoids. Potential mechanisms of their deleterious effects on bone. J Clin Invest 102 (2) : 274-82. 6. Lane NE, Sanchez S, Modin GW, Genant HK, Pierini E, Arnaud CD 1998 Parathyroid hormone treatment can reverse corticostero id-induced osteoporosis. Results of a randomized controlled clinical trial. J Clin Invest 102 (8) : 1627-33. 7. Alexander JM, Bab I, Fish S, Muller R, Uchiyama T, Gronowicz G, Nahounou M, Zhao Q, White DW, Chorev M, Gazit D, Rose nblatt M 2001 Human parathyroid hormone 1-34 reverses bone loss in ovariectomized mice. J Bone Miner Res 16 (9) : 1665-73. 8. Zhou H, Iida-Klein A, Lu SS, Ducayen-Knowles M, Levine LR, Dempster DW, Lindsay R 2003 Anabolic action of parathyroid hormone on cortical and cancellous bone differs between axial and appendicular skeletal sites in mice. Bone 32 (5) : 513-20. 9. Braith RW, Magyari PM, Fulton MN, Ara nda J, Walker T, Hill JA 2003 Resistance exercise training and alendronate reverse gl ucocorticoid-induced osteoporosis in heart transplant recipients. J Heart Lung Transplant 22 (10) : 1082-90. 10. Braith RW, Mills RM, Welsch MA, Kelle r JW, Pollock ML 1996 Resistance exercise training restores bone mineral density in hear t transplant recipients. J Am Coll Cardiol 28 (6) :1471-7. 11. Boulos P, Ioannidis G, Adachi JD 20 00 Glucocorticoid-induced osteoporosis. Curr Rheumatol Rep 2 (1) : 53-61.

PAGE 112

112 12. Clarke B, Leidig-Bruckner G 2005 Fracture Prevalence and Incidence in Solid Organ Transplant Recipients. In: Compston J, Shane E (eds.) Bone Disease of Organ Transplantation. El Sevier Academic Press, Boston. 13. Dennison E, Cooper C 2002 Epidemiology of glucocorticoid-induced osteoporosis. Front Horm Res 30: 121-6. 14. Rehman Q, Lane NE 2003 Effect of glucoc orticoids on bone density. Med Pediatr Oncol 41 (3) : 212-6. 15. Sambrook P, Lane NE 2001 Corticosteroid os teoporosis. Best Pract Res Clin Rheumatol 15 (3) : 401-13. 16. Boling EP 2004 Secondary osteoporosis: underlying disease and the risk for glucocorticoid-induced osteoporosis. Clin Ther 26 (1) : 1-14. 17. Cohen A, Shane E 2003 Osteoporosis after so lid organ and bone ma rrow transplantation. Osteoporos Int 14 (8) : 617-30. 18. Cohen D, Adachi JD 2004 The treatment of glucocorticoid-induced osteoporosis. J Steroid Biochem Mol Biol 88 (4-5) : 337-49. 19. Rodino MA, Shane E 1998 Osteoporosis af ter organ transplantation. Am J Med 104 (5) : 459-69. 20. Shane E, Epstein S 2001 Transplantati on Osteoporosis. Tran splantation Reviews 15 (1) : 11-32. 21. Dalle Carbonare L, Arlot ME, Chavassie ux PM, Roux JP, Portero NR, Meunier PJ 2001 Comparison of trabecular bone microarchite cture and remodeling in glucocorticoidinduced and postmenopausal oste oporosis. J Bone Miner Res 16 (1) : 97-103. 22. McIlwain HH 2003 Glucocorticoid-induced osteoporosis: pathogenesis, diagnosis, and management. Prev Med 36 (2) : 243-9. 23. Lafage-Proust MH, Boudignon B, Thomas T 2003 Glucocorticoid-induced osteoporosis: pathophysiological data and recent treatments. Joint Bone Spine 70 (2) : 109-18. 24. Braith RW, Magyari PM, Fulton MN, Lisor CF, Vogel SE, Hill JA, Aranda JM Comparison of calcitonin versus calcitonin + resistance exercise as prophylaxis for osteoporosis in heart transplant recipients. Transplantation In Press 25. Braith RW, S.D. G, Musto T, Mitchell MJ, Baz MA 1998 Resistance exercise restored bone mineral density in an osteoporotic patie nt before lung transp lantation. Journal of Cardiopulmonary Rehabilitation 36: 18-23.

PAGE 113

113 26. Mitchell M, Fulton M, Lisor C, Baz M, Br aith R 2002 Resistance training attenuates glucocorticoid-induced osteoporosis in lung tr ansplant recipients. Journal of Heart and Lung Transplantation in review 27. Uusi-Rasi K, Sievanen H, Heinonen A, Kannus P, Vuori I 2004 Effect of discontinuation of alendronate treatment a nd exercise on bone mass and physical fitness: 15-month follow-up of a randomized, controlled trial. Bone 35 (3) : 799-805. 28. Mitlak BH 2002 Parathyroid hormone as a therapeutic agent. Curr Opin Pharmacol 2 (6) : 694-9. 29. Neer RM, Arnaud CD, Zanchetta JR, Prince R, Gaich GA, Reginster JY, Hodsman AB, Eriksen EF, Ish-Shalom S, Genant HK, Wang O, Mitlak BH 2001 Effect of parathyroid hormone (1-34) on fractures and bone minera l density in postmenopausal women with osteoporosis. N Engl J Med 344 (19) : 1434-41. 30. Rittmaster RS, Bolognese M, Ettinger MP Hanley DA, Hodsma n AB, Kendler DL, Rosen CJ 2000 Enhancement of bone mass in osteoporotic women with parathyroid hormone followed by alendronate. J Clin Endocrinol Metab 85 (6) : 2129-34. 31. Kaufman JM, Orwoll E, Goemaere S, San Mart in J, Hossain A, Dalsky GP, Lindsay R, Mitlak BH 2005 Teriparatide effects on verteb ral fractures and bone mineral density in men with osteoporosis: treatment and disc ontinuation of therapy. Osteoporos Int 16 (5) : 510-6. 32. Finkelstein JS, Hayes A, Hunzelman JL, Wyla nd JJ, Lee H, Neer RM 2003 The effects of parathyroid hormone, alendrona te, or both in men with osteoporosis. N Engl J Med 349 (13) : 1216-26. 33. Body JJ, Gaich GA, Scheele WH, Kulkarni PM, Miller PD, Peretz A, Dore RK, CorreaRotter R, Papaioannou A, Cumming DC, H odsman AB 2002 A randomized double-blind trial to compare the efficacy of teriparatide [recombinant human parathyroid hormone (134)] with alendronate in postmenopausal wo men with osteoporosis. J Clin Endocrinol Metab 87 (10) : 4528-35. 34. Lindsay R, Nieves J, Formica C, Hennema n E, Woelfert L, Shen V, Dempster D, Cosman F 1997 Randomised controlled study of effect of parathyroid hormone on vertebral-bone mass and fracture inciden ce among postmenopausal women on oestrogen with osteoporosis. Lancet 350 (9077) : 550-5. 35. Pettway GJ, Schneider A, Koh AJ, Widjaj a E, Morris MD, Meganck JA, Goldstein SA, McCauley LK 2005 Anabolic actions of PTH (1-34): use of a novel tissue engineering model to investigate tem poral effects on bone. Bone 36 (6) : 959-70. 36. Binz K, Schmid C, Bouillon R, Froesch ER, Jurgensen K, Hunziker EB 1994 Interactions of insulin-like growth factor I with de xamethasone on trabecular bone density and mineral metabolism in rats. Eur J Endocrinol 130(4) : 387-93.

PAGE 114

114 37. King CS, Weir EC, Gundberg CW, Fox J, Insogna KL 1996 Effects of continuous glucocorticoid infusion on bone metabo lism in the rat. Calcif Tissue Int 59 (3) : 184-91. 38. Shen V, Birchman R, Liang XG, Wu DD, Lindsay R, Dempster DW 1997 Prednisolone alone, or in combination with estrogen or dietary calcium deficiency or immobilization, inhibits bone formation but does not i nduce bone loss in mature rats. Bone 21 (4) : 345-51. 39. Manolagas SC, Weinstein RS 1999 New develo pments in the pathogenesis and treatment of steroid-induced osteoporosis. J Bone Miner Res 14 (7) : 1061-6. 40. Mosekilde L 1995 Assessing bone quality--ani mal models in preclinical osteoporosis research. Bone 17 (4 Suppl) : 343S-352S. 41. Thompson DD, Simmons HA, Pirie CM, Ke HZ 1995 FDA Guidelines and animal models for osteoporosis. Bone 17 (4 Suppl) : 125S-133S. 42. Weinstein RS 2001 Glucocorticoid-induced osteoporosis. Rev Endocr Metab Disord 2 (1) : 65-73. 43. Weinstein RS, Manolagas SC 2000 Apoptosis and osteoporosis. Am J Med 108 (2) : 15364. 44. Lane NE, Sanchez S, Modin GW, Genant HK, Pierini E, Arnaud CD 2000 Bone mass continues to increase at the hip after para thyroid hormone treatment is discontinued in glucocorticoid-induced osteoporos is: results of a randomized c ontrolled clinical trial. J Bone Miner Res 15 (5) : 944-51. 45. Iida-Klein A, Zhou H, lu SS, levine LR, Ducayen-Knowles M, Dempster D, Nieves J, lindsay R 2002 Anabolic action of parathyroid hormone is skeletal site specific at the tissue and cellular levels in mice. J ournal of Bone and Mineral Research 17 (5) : 808-816. 46. ALA July 2005 Estimated prevalence and in cidence of lung disease by lung association territory American Lung Associ ation: Epidemiology and st atistical unit research and program services, pp 1-53. 47. 2005 Organ Procurement andTransplanta tion Network Statistical Database. 48. Swarthout JT, D'Alonzo RC, Selvamurugan N, Partridge NC 2002 Parathyroid hormonedependent signaling pathways regul ating genes in bone cells. Gene 282 (1-2) : 1-17. 49. Wronski TJ, Yen C-F, Qi H, Dann LM 1993 Pa rathyroid hormone is more effective than estrogen or bisphosphonates fo r restoration of lost bone mass in ovariectomized rats. Endocrinology 132 (2) : 823-831. 50. Li M, Shen Y, Halloran BP, Baumann BD, Miller K, Wronski TJ 1996 Skeletal response to corticosteroid deficiency and excess in growing male rats. Bone 19 (2) : 81-8.

PAGE 115

115 51. Parfitt AM, Drezner MK, Glorieux FH, Kani s JA, Malluche H, Meunier PJ, Ott SM, Recker RR 1987 Bone histomorphometry: standard ization of nomenclature, symbols, and units. Report of the ASBMR Histomorphometry Nomenclature Committee. J Bone Miner Res 2 (6) : 595-610. 52. Conover WJ 1980 Practical Nonparametric St atistics. Wiley and Sons, New York, NY, pp 229-237. 53. Bonner F, Worrell RV 1991 A Basic Science Primer in Orthopaedics. 54. Reeve J 2000 How do women develop fragile bones? J Steroid Biochem Mol Biol 74 (5) : 375-81. 55. Kanis J 1991 Calcium Requirements for Optim al Skeletal Health in Women. Calcified Tissue International Supplement 49: S33-S41. 56. Parfitt AM 1992 Implications of Architect ure for the Pathogenesis and Prevention of Vertebral Fracture. bone 13 (Supplement) : S41-S47. 57. Vaananen HK, Zhao H, Mulari M, Halle en JM 2000 The Cell Biology of Osteoclast Function. J Cell Sci 113: 377-381. 58. Arita S, Ikeda S, Sakai A, Okimoto N, Ak ahoshi S, Nagashima M, Nishida A, Ito M, Nakamura T 2004 Human parathyroid hormone (1-34) increases mass and structure of the cortical shell, with resultant increase in lumbar bone strength, in ovariectomized rats. J Bone Miner Metab 22 (6) : 530-40. 59. Vaananen HK, Zhao H, Mulari M, Halle en JM 2000 The cell biology of osteoclast function. J Cell Sci 113 ( Pt 3): 377-81. 60. Hofbauer LC, Khosla S, Dunstan CR, Lacey DL, Boyle WJ, Riggs BL 2000 The roles of osteoprotegerin and osteoprotegerin ligand in the paracrine regulat ion of bone resorption. J Bone Miner Res 15 (1) : 2-12. 61. Martin RB 2000 Toward a Unifying Theory of Bone Remodeling. Bone 26 (1). 62. Martin RB, Burr DB, Sharkey NA 1998 Skelet al Tissue Mechanics. Springer-Verelag, New York. 63. Notelovitz M, Martin D, Tesar R 1991 Estr ogen Therapy and Variable-Resistance weight training increase bone mineral in surgically menopausal women. Journal of Bone and Mineral Research 6: 583-590. 64. Hazelwood SJ, Bruce Martin R, Rashid MM, Rodrigo JJ 2001 A mechanistic model for internal bone remodeling exhibits different dynamic responses in disuse and overload. J Biomech 34 (3) : 299-308.

PAGE 116

116 65. Takahashi N, Udagawa N, Suda T 1999 A ne w member of tumor n ecrosis factor ligand family, ODF/OPGL/TRANCE/RANKL, regulates osteoclast differentiation and function. Biochem Biophys Res Commun 256 (3) : 449-55. 66. Frost H 1997 On Our Age-Related Bone Loss: Insights from a New Paradigm. Journal of Bone and Mineral Research 12 (10). 67. Bronner F, Worrell RV 1991 A Basic Science Primer in Orthopaedics. Lippincott, WIlliams, and Wilkins, New York. 68. Hazelwood SJ, Martin RB, Rashid MM, Rodrigo J 2001 A Mechanistic Model for Internal Bone Remodeling Exhibits Different Dynamic Responses in Disuse and Overload. Journal of Biomechanics 34: 299-308. 69. Kostenuik PJ 2005 Osteoprotegerin and RANKL regulate bone resorption, density, geometry and strength. Curr Opin Pharmacol 5 (6) : 618-25. 70. Oh ES, Rhee EJ, Oh KW, Lee WY, Baek KH, Yoon KH, Kang MI, Yun EJ, Park CY, Choi MG, Yoo HJ, Park SW 2005 Circulating os teoprotegerin levels are associated with age, waist-to-hip ratio, seru m total cholesterol, and lowdensity lipoprotein cholesterol levels in healthy Korean women. Metabolism 54 (1) : 49-54. 71. Sandy J, Davies M, Prime S, Farndale R 1998 Signal pathways that transduce growth factor-stimulated mitogene sis in bone cells. Bone 23 (1) : 17-26. 72. Dalle Carbonare L, Bertoldo F, Valenti MT, Ze nari S, Zanatta M, Sella S, Giannini S, Cascio VL 2005 Histomorphometric analysis of glucocorticoid-induced osteoporosis. Micron 36 (7-8) : 645-52. 73. Dalle Carbonare L, Chavassieux PM, Arlot ME, Meunier PJ 2002 Bone histomorphometry in untreated and treated gl ucocorticoid-induced osteoporosis. Front Horm Res 30: 37-48. 74. Patschan D, Loddenkemper K, Buttg ereit F 2001 Molecular mechanisms of glucocorticoid-induced osteoporosis. Bone 29 (6) : 498-505. 75. Reid IR 2000 Glucocorticoid-induced oste oporosis. Baillieres Be st Pract Res Clin Endocrinol Metab 14 (2) : 279-98. 76. Manelli F, Giustina A 2000 Glucocorticoid -induced osteoporosis. Trends Endocrinol Metab 11 (3) : 79-85. 77. Adcock IM 2004 Corticosteroids: limitations and future prospects for treatment of severe inflammatory disease. Drug Discovery Today: Therapeutic Strategies 1 (3) : 321-328. 78. Demoly P, Chung KF 1998 Pharmacol ogy of corticosteroids. Respir Med 92(3) : 385-94.

PAGE 117

117 79. Schacke H, Docke WD, Asadullah K 2002 Mech anisms involved in the side effects of glucocorticoids. Pharmacol Ther 96 (1) : 23-43. 80. Canalis E 2003 Mechanisms of glucocorti coid-induced osteoporosis. Curr Opin Rheumatol 15 (4) : 454-7. 81. Canalis E, Bilezikian JP, Angeli A, Gius tina A 2004 Perspectives on glucocorticoidinduced osteoporosis. Bone 34 (4) : 593-8. 82. Ton FN, Gunawardene SC, Lee H, Neer RM 2005 Effects of low-dose prednisone on bone metabolism. J Bone Miner Res 20 (3) : 464-70. 83. Boyde A, Maconnachie E, Reid SA, Delling G, Mundy GR 1986 Scanning electron microscopy in bone pathology: review of me thods, potential and applications. Scan Electron Microsc (Pt 4) : 1537-54. 84. Tamura Y, Okinaga H, Takami H 2004 Gluc ocorticoid-induced osteoporosis. Biomed Pharmacother 58 (9) : 500-4. 85. Malyszko J, Malyszko JS, Wolczynski S, Mysliwiec M 2003 Osteoprotegerin and its correlations with new markers of bone forma tion and bone resorption in kidney transplant recipients. Transplant Proc 35 (6) : 2227-9. 86. Ferrari P 2003 Cortisol and the renal handli ng of electrolytes: role in glucocorticoidinduced hypertension and bone disease. Best Pract Res Clin Endocrinol Metab 17 (4) : 57589. 87. Canalis E 1996 Clinical review 83: Mech anisms of glucocorticoid action in bone: implications to glucocorticoid-induced osteoporosis. J Clin Endocrinol Metab 81 (10) : 3441-7. 88. Sivagurunathan S, Muir MM, Brennan TC Seale JP, Mason RS 2005 Influence of glucocorticoids on human osteoclast ge neration and activity. J Bone Miner Res 20 (3) : 390-8. 89. Reid DM, Harvie J 1997 Secondary oste oporosis. Baillieres C lin Endocrinol Metab 11 (1) : 83-99. 90. Brixen KT, Christensen PM, Ejersted C, Langdahl BL 2004 Teriparatide (biosynthetic human parathyroid hormone 1-34): a new para digm in the treatment of osteoporosis. Basic Clin Pharmacol Toxicol 94 (6) : 260-70. 91. Fox J 2002 Developments in parathyroid horm one and related peptides as bone-formation agents. Curr Opin Pharmacol 2 (3): 338-44. 92. Debiais F 2003 Efficacy data on teriparatide (parathyroid hormone) in patients with postmenopausal osteoporosis. Joint Bone Spine 70 (6) : 465-70.

PAGE 118

118 93. Selye H 1932 On the stimulation of new bone formation with parathyroid extract and irradiated ergosterol. Endocrinology 16: 547-558. 94. Fitzpatrick LA, Bilezikian JP 1996 Actions of Parathyroid Hormone. In: Bilezikian JP, Raisz LG, Rodan GA (eds.) Principles of Bone Biology. Academic Press, San Diego. 95. Locklin RM, Khosla S, Turner RT, Riggs BL 2003 Mediators of the biphasic responses of bone to intermittent and continuously administered parathyroid hormone. J Cell Biochem 89 (1) : 180-90. 96. Canalis E, Centrella M, Burch W, Mc Carthy TL 1989 Insulin-lik e growth factor I mediates selective anabolic effects of parath yroid hormone in bone cultures. J Clin Invest 83 (1) : 60-5. 97. Rosen CJ 2004 What's new with PTH in oste oporosis: where are we and where are we headed? Trends Endocrinol Metab 15 (5) : 229-33. 98. Gensure RC, Gardella TJ, Juppner H 2005 Parathyroid hormone and parathyroid hormone-related peptide, and their re ceptors. Biochem Biophys Res Commun 328 (3) : 666-78. 99. Rosen CJ, Bilezikian JP 2001 Clinical review 123: Anabolic ther apy for osteoporosis. J Clin Endocrinol Metab 86 (3) : 957-64. 100. Quattrocchi E, Kourlas H 2004 Teri paratide: a review. Clin Ther 26 (6) : 841-54. 101. Onyia JE, Helvering LM, Gelbert L, Wei T, Huang S, Chen P, Dow ER, Maran A, Zhang M, Lotinun S, Lin X, Halladay DL, Miles RR, Kulkarni NH, Ambrose EM, Ma YL, Frolik CA, Sato M, Bryant HU, Turner RT 2005 Molecular profile of catabolic versus anabolic treatment regimens of parathyroi d hormone (PTH) in ra t bone: an analysis by DNA microarray. J Cell Biochem 95 (2) : 403-18. 102. Xing L, Boyce BF 2005 Regulation of apoptosis in osteoclasts and osteoblastic cells. Biochem Biophys Res Commun 328 (3) : 709-20. 103. Tashjian AH, Jr., Chabner BA 2002 Commentary on clinical safety of recombinant human parathyroid hormone 1-34 in the treatment of osteoporosis in men and postmenopausal women. J Bone Miner Res 17 (7) : 1151-61. 104. Vahle JL, Long GG, Sandusky G, Westmore M, Ma YL, Sato M 2004 Bone neoplasms in F344 rats given teriparatide [rhPTH(1-34)] are dependent on duration of treatment and dose. Toxicol Pathol 32 (4) : 426-38. 105. Vahle JL, Sato M, Long GG, Young JK, Francis PC, Engelhardt JA, Westmore MS, Linda Y, Nold JB 2002 Skeletal changes in ra ts given daily subcutan eous injections of recombinant human parathyroid hormone (134) for 2 years and relevance to human safety. Toxicol Pathol 30 (3) : 312-21.

PAGE 119

119 106. Berg C, Neumeyer K, Kirkpatrick P 2003 Teriparatide. Nat Rev Drug Discov 2 (4) : 257-8. 107. U.S.Forteo 2004 Prescribing information, Eli Lilly and Company. 108. Qin L, Raggatt LJ, Partridge NC 2004 Pa rathyroid hormone: a double-edged sword for bone metabolism. Trends Endocrinol Metab 15 (2) : 60-5. 109. Audran M, Insalaco P 2003 Parathyroid hor mone therapy for osteoporosis. Joint Bone Spine 70 (5) : 315-7. 110. Jiang Y, Zhao JJ, Mitlak BH, Wang O, Genant HK, Eriksen EF 2003 Recombinant human parathyroid hormone (1-34) [teriparat ide] improves both cortical and cancellous bone structure. J Bone Miner Res 18 (11) : 1932-41. 111. Dempster DW, Cosman F, Kurland ES, Zhou H, Nieves J, Woelfert L, Shane E, Plavetic K, Muller R, Bilezikian J, Lindsay R 2001 E ffects of daily treatment with parathyroid hormone on bone microarchitecture and turnover in patients with osteoporosis: a paired biopsy study. J Bone Miner Res 16 (10) : 1846-53. 112. Adams AE, Rosenblatt M, Suva LJ 1999 Iden tification of a novel parathyroid hormoneresponsive gene in human osteoblastic cells. Bone 24 (4) : 305-13. 113. Samuels A, Perry MJ, Gibson RL, Colley S, Tobias JH 2001 Role of endothelial nitric oxide synthase in estrogeninduced osteogenesis. Bone 29 (1) : 24-9. 114. Weinstein RS, Jia D, Powers CC, Stewart SA, Jilka RL, Parfitt AM, Manolagas SC 2004 The skeletal effects of glucocorticoid exce ss override those of orchidectomy in mice. Endocrinology 145 (4) : 1980-7. 115. Oxlund H, Ortoft G, Thomsen JS, Daniel sen CC, Ejersted C, Andreassen TT 2006 The anabolic effect of PTH on bone is attenuate d by simultaneous glucocorticoid treatment. Bone 39 (2) : 244-52. 116. Samuels A, Perry MJ, Gibson R, Tobias JH 2001 Effects of combination therapy with PTH and 17beta-estradiol on long bones of female mice. Calcif Tissue Int 69 (3) : 164-70. 117. Sibonga JD, Iwaniec UT, Shogren KL, Rosen CJ, Turner RT 2007 Effects of parathyroid hormone (1-34) on tibia in an adult rat model for chronic alcohol abuse. Bone 40 (4) : 1013-20. 118. Knopp E, Troiano N, Bouxsein M, Sun BH, Lo stritto K, Gundberg C, Dziura J, Insogna K 2005 The effect of aging on the skeletal response to intermittent treatment with parathyroid hormone. Endocrinology 146 (4) : 1983-90. 119. Halloran BP, Ferguson VL, Simske SJ, Burghardt A, Venton LL, Majumdar S 2002 Changes in bone structure and mass with adva ncing age in the male C57BL/6J mouse. J Bone Miner Res 17 (6) : 1044-50.

PAGE 120

120 120. Jilka RL, Weinstein RS, Takahashi K, Parfitt AM, Manolagas SC 1996 Linkage of decreased bone mass with impaired osteoblas togenesis in a murine model of accelerated senescence. J Clin Invest 97 (7) : 1732-40. 121. Turner RT, Evans GL, Lotinun S, Lapke PD, Iwaniec UT, Morey-Holton E 2007 Doseresponse effects of intermittent PTH on cancellous bone in hindlimb unloaded rats. J Bone Miner Res 22 (1) : 64-71. 122. Turner RT, Lotinun S, Hefferan TE, MoreyHolton E 2006 Disuse in adult male rats attenuates the bone anabolic response to a therap eutic dose of parathyroid hormone. J Appl Physiol 101 (3) : 881-6.

PAGE 121

121 BIOGRAPHICAL SKETCH Kathleen S. Howe was commissi oned as an officer in the United States Air Force after her graduation from the University of Central Florid a in 1978. She spent the next 20 years serving in a number of positions in the United States Air For ce before retiring as a Lieutenant Colonel in 2000. After her retirement, she returned to the academic world to pursu e her interest in bone metabolism. She has spent the past 7 years conc entrating on her studies in exercise physiology and bone biology. Her research has centered on inte rventions to reverse the effects of aging and disease on bone and reversing th e effects of secondary osteopor osis. She plans to continue research related to bone metabolism in the futu re. She has accepted a postdoctoral position in the Department of Nutrition and Exercise Scie nce at Oregon State University where she will examine the effects of alcohol on bo ne structure and metabolism.