AD_________________

Award Number: W81XWH-09-2-0122 TITLE: Amelioration of Ischemia/Reperfusion Injury During Resuscitation from Hemorrhage by Induction of Heme Oxygenase-1 (HO-1) in a Conscious Mouse Model of Uncontrolled Hemorrhage PRINCIPAL INVESTIGATOR: Phillip Bowman CONTRACTING ORGANIZATION: The Geneva Foundation

Tacoma, WA 98402 REPORT DATE: October 2013 TYPE OF REPORT: Addendum to Final PREPARED FOR: U.S. Army Medical Research and Materiel Command Fort Detrick, Maryland 21702-5012 DISTRIBUTION STATEMENT: Approved for Public Release; Distribution Unlimited The views, opinions and/or findings contained in this report are those of the author(s) and should not be construed as an official Department of the Army position, policy or decision unless so designated by other documentation.

REPORT DOCUMENTATION PAGE Form Approved

OMB No. 0704-0188 Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing this collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden to Department of Defense, Washington Headquarters Services, Directorate for Information Operations and Reports (0704-0188), 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to any penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number. PLEASE DO NOT RETURN YOUR FORM TO THE ABOVE ADDRESS.

1. REPORT DATE October 2013

2. REPORT TYPE Addendum to Final

3. DATES COVERED 1 September 2012 – 31 August 2013

4. TITLE AND SUBTITLE

Amelioration of Ischemia/Reperfusion Injury During Resuscitation from Hemorrhage by Induction of Heme Oxygenase-1 (HO-1) in a Conscious Mouse Model of Uncontrolled Hemorrhage

5a. CONTRACT NUMBER

5b. GRANT NUMBER

W81XWH-09-2-0122

5c. PROGRAM ELEMENT NUMBER

6. AUTHOR(S)

Phillip Bowman 5d. PROJECT NUMBER

5e. TASK NUMBER

Email:phillip.d.bowman4.civ@mail.mil

5f. WORK UNIT NUMBER 7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES)

The Geneva Foundation AND ADDRESS(ES)

8. PERFORMING ORGANIZATION REPORT NUMBER

Tacoma, WA 98402

9. SPONSORING / MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR’S ACRONYM(S) U.S. Army Medical Research and Materiel Command

Fort Detrick, Maryland 21702-5012

11. SPONSOR/MONITOR’S REPORT

NUMBER(S)

12. DISTRIBUTION / AVAILABILITY STATEMENT

Approved for Public Release; Distribution Unlimited 13. SUPPLEMENTARY NOTES

14. ABSTRACT Ischemia occurs whenever there is interruption of the flow of blood to tissues or organs. It is the most common cause of death in heart disease and stroke as well as traumatic injury. Survival of the initial insult is followed by further injury that occurs during the reintroduction of oxygen with the restoration of blood flow. This injury occurs following hemorrhage because some tissues are deprived of blood to protect others as part of the fight or flight response. Heme oxygenase-1 (HO-1) induction is correlated with a significant reduction in ischemic injury and 1-[2cyano-3, 12-dioxoleana-1, 9(11)-dien 28-oy}limidazole (CDDO-lm) a new synthetic triterpenoid that has been shown to possess potent anti-inflammatory and antioxidant properties, and is a potent inducer of HO-1. The hypothesis to be tested is that a significant reduction in indicies of I/R injury will be obtained by induction of HO-1 during resuscitation with CDDO-IM following hemorrhage.

15. SUBJECT TERMS

Hemorrhage shock, cytoprotection, ischemia/reperfusion injury, drugs 16. SECURITY CLASSIFICATION OF:

17. LIMITATION OF ABSTRACT

18. NUMBER OF PAGES

19a. NAME OF RESPONSIBLE PERSON USAMRMC

a. REPORT

U b. ABSTRACT

U c. THIS PAGE

U

UU 27

19b. TELEPHONE NUMBER (include area

code)

Table of Contents

Page

Introduction ...................................................................................... 4

Body ............................................................................................... 4

Key Research Accomplishments .... ..... ... .......... ....... . ...... ... .. . .. ... .......... 5

Reportable Outcomes ... ........ ... .................. . ...... .............................. .. . 6

Conclusion ............ . ......... . ........... ...................... .. .. ..... .. ... . .. ... ... ... .. .. . 7

References ..... .. .............. . ........... .. .. ..... .. . .. ...... ........... . .............. ......... 8

Appendices ......... . ... ....................... ............. . ...... . ...... . ...... . .... ........ .. .. 8

INTRODUCTION: HIF1a, a master regulator of the hypoxic response has been implicated in ischemic preconditioning. lschemic preconditioning has been shown to provide significant protection from a subsequent lethal ischemic event. Additionally, Heme oxygenase-1 (H0-1) is an inducible Phase 2 enzyme that degrades toxic heme. Heme contains an iron and when released under pathological conditions such as cellular stresses and ischemia, free heme may act as a source of free radicals. Cells have therefore evolved a system to degrade heme, a system composed of inducible heme oxygenases 1 (H0-1) and constitutive H0-2. The end products of the degradation include cytoprotective biliverdin and carbon monoxide; as a result, heme oxygenases are potentially cytoprotective. lschemia occurs whenever there is interruption of the flow of blood to tissues or organs. It is the most common cause of death in heart disease and stroke as well as traumatic injury. Survival of the initial insult is followed by further injury that occurs during the reintroduction of oxygen with the restoration of blood flow. This injury occurs following hemorrhage because some tissues are deprived of blood to protect others as part of the fight or flight response. Recent investigations have shown 2-cyano-3, 12 dioxooleana-1,9 dien-28-oyl imidazoline (CDDO-lm) a new synthetic triterpenoid to possess potent anti-inflammatory and antioxidant properties, and is a potent inducer of H0-1. We hypothesized that chemically induced H0-1 upregulation with the novel triterpenoid CDDO-lm (2-cyano-3, 12 dioxooleana-1 ,9 dien-28-oyl imidazoline), a robust inducer of Phase 2 genes, protects against ischemic injury. To measure cytoprotection in terms of luminescence, we also screened genetically engineered mouse cells that express luciferase when HIF1 a accumulates.

BODY: CDDO-lm is a synthetic triterpenoid recently shown to induce cytoprotective genes through the Nrf2-Keap1 pathway, an important mechanism for the induction of cytoprotective genes in response to oxidative stress. Heme oxgenase-1 is highly inducible and its induction is correlated with significant protection from the deleterious effects of ischemia. CDDO-lm (2-cyano-3, 12 dioxooleana-1 ,9 dien-28-oyl imidazoline), a new synthetic triterpenoid has been shown to possess potent anti-inflammatory and antioxidant properties, and is a potent inducer of H0-1 and is being investigated as an additive to a new resuscitation fluid that might being counteracting the deleterious effects of the ischemia of hemorrhage shortly after injury during the initial resuscitation. One of the most critical components of developing a new drug of treatment of a specific disease state is determination of an appropriate dose of the drug with maximum benefit and minimum off­target effects. We employed a new technique, termed snapshot pharmacokinetics, to hone in on an appropriate dose of CDDO-lm for use in the mouse model of hemorrhage. Following determination of an appropriate dose we determined the timing of resuscitation for maximum benefit of the drug.

4

We also tested and screened other drugs such as CAPE, CAPA for their ability to induce H0-1 and HIF1 a and produce a cytoprotective effect. We screened genetically engineered mouse cells that express luciferase when HIF1 a accumulates. Deferoxamine induces HIF1a by inhibiting the activity of Fe++ dependent prolyl hydroxylase which is required for activation of the oxygen dependent domain of HIF1 a and was used as a positive control. Caffeic acid phenethyl ester (CAPE) has been previously suggested to inhibit HIF1 a prolyl hydroxylase. We have found that Caffeic acid phenethyl amide (CAPA) and CAPE along with CDDO-lm, which induced H0-1 mediated cytoprotection against menadione-induced-oxidative stress, also induces HIF1a and this may explain their cytoprotective effect.

KEY RESEARCH ACCOMPLISHMENTS:

Due to Base Realignment and Closing (BRAC) issues, significant delays in research occurred at two times during the study. Firstly, the USAISR vivarium was closed for about two years in order to renovate it and merge it with the new vivarium in the newly constructed Battlefield Health and Trauma Center for Excellence. During th is period animals were housed at the animal facility at Brook City Base, San Antonio, TX. However, in vivo imaging still had to be performed at the USAISR where the Xenogen in vivo imaging system resided. This required transporting the animals from Brooks City Base to USAISR (about 15 miles away). Also , the rules during this period were that no animal bought from Brooks City Base could be in the Institute for more than 8 hr. Consequently, the imaging data obtained during th is period could not be used as the behavior of the mice was significantly when this additional stress of transportation was factored in. Secondly, the entire OCR (Damage Control, Resuscitation) group at US Army Institute of Surgical Research (USAISR) was moved to a new facility in Oct. of 2011 . This delayed analysis mouse tissues obtained from repeats of the work done at Brooks City Base an add itional 6 months. In addition, a postdoctoral fellow working on an MRMC project left in August 2010 for a teaching job and the postdoctoral fellow working under the Geneva Foundation project had to fill the MRMC postdoctoral fellow position as MRMC projects pays for 90% of the research budget and takes precedence over congressional projects. I was finally able to hire a post doctoral fellow in August 2012 to work full time on this project.

Jan-Dec 2009: Preliminary Studies-> Studied Structure activity relationship of Caffeic Acid Phenethyl Ester (CAPE) and its amide derivative CAPA against oxidant stress in human endothelial cells; Used non-invasing imaging techniques as a tool to demonstrate hemorrhage-induced global ischemia with a transgenic mouse expressing luciferase coupled to hypoxia-inducible factor (HIF1a).

Jan-Mar 2010: Established role for hypoxia in some organs of the mouse following hemorrhage of the FVB.1 29S6-Gt(ROSA)26Sortml (HIF1 aluc)Kael/J (HIF1 a luc) inbred

5

strain using luminometry analysis. The intestine, spleen and liver were effected organ while brain , lung, skeletal muscle and heart were not much affected.

Jan-June 2010: Determined that 100 nm CDDO was optimal in vitro for induction of H0-1 in the skin fibroblasts of the HIF1 aluc strain of mouse.

Mar-Aug 2010: Determined that 100 nM CDDO was also effective in inducing H0-1 in human umbilical vein endothelial (HUVEC) cells indicating a cross species benefit.

August 2010 - October 2011 : Compared CDDO-lm, CAPE, CAPA induced H0-1 mediated cytoprotection against menadione-induced-oxidative stress in HUVEC cells.

November 2011 - December 2012: Determined appropriate dose of CDDO-lm (50-100 nm) in a mouse model.

January 2012 -August 2013: Evaluated cytoprotective effects of CDDO-lm in a mouse model. Analyzed various organs, using Western Blot and other biochemical assays, for proof of cytoprotection.

REPORTABLE OUTCOMES:

List of Presentations and Manuscripts. The presentations and manuscript contains all the relevant data (figures, tables, conclusions) pertaining to this research . Copies of manuscript and presentations are attached with this report.

PRESENTATIONS:

1. Experimental Biology 2009: Cytoprotective effect of a synthetic triterpenoid against oxidative stress in human umbilical vein endothelial cells (HUVEC). FASEB J. April 2009 23 (Meeting Abstract Supplement) 937.7

2. ATACC 2009: Noninvasive imaging of hemorrhage-induced global ischemia with a transgenic mouse expressing luciferase coupled to hypoxia-inducible factor (HIF1 a).

3. Experimental Biology 2009: Structure activity relationship of Caffeic Acid Phenethyl Ester (CAPE) and its amide derivative CAPA against oxidant stress in human endothelial cells. FASEB J. April 2009 23 (Meeting Abstract Supplement) 937.8.

4. Experimental Biology 2010: Cytoprotection of Human Endothelial Cells from Oxidative Stress by Polyphenols: the Role of Gene Expression versus Direct Antioxidant Effect.

5. Experimental Biology 2010: Induction of Hypoxia Inducible Factor 1 Alpha (HIF1a) by Caffeic Acid Phenethyl Ester (CAPE) and Caffeic Acid Phenethyl Amide (CAPA) in Mouse Skin Fibroblasts. FASEB J. April 2010 24 (Meeting Abstract Supplement) 760.2.

6

6. ATACC 2011: Poor Correlation between In Vivo Imaging and Production of Light by Organs in Transgenic Mouse Engineered to Express Luciferase in Response to Hypoxia.

7. Experimental Biology 2011 : Time Course and Network Analysis of 1-[2-Cyano-3, 12-dioxooleana-1,9(11 )-dien-28-oyl]imidizole (CDDO-lm) Induction of Cytoprotective Genes in Human Umbilical Vein Endothelial Cells (HUVEC) Against Oxidant Stress. FASEB J. April 2011 25 (Meeting Abstract Supplement) 1090.3.

8. American Association of Pharmaceutical Scientists (AAPS) 2012: Determination of the Minimum Exposure Time for Effecting Cytoprotection in Human Umbilical Vein Endothelial Cells (HUVEC) for Caffeic Acid Phenylethyl Ester (CAPE) and Amide (CAPA).

9. American Association of Pharmaceutical Scientists (AAPS) 2013: Comparison of atmospheric oxygen versus physiological levels on cytotoxicity of menadione and cytoprotection by antioxidants in human endothelial cells.

10. American Association of Pharmaceutical Scientists (AAPS) 2013: Pharmacokinetic Profiles of Caffeic Acid Phenethyl Amide (CAPA) and Caffeic Acid Phenethyl Ester (CAPE) in Male Sprague-Dawley Rats.

11 .American Association of Pharmaceutical Scientists (AAPS) 2013: Comparison of caffeic acid phenylester (CAPE), caffeic acid phenylamide (CAPA) and 2-cyano-3, 12 dioxooleana-1,9 dien-28-imidazolide (CDDO-lm) in protecting human endothelial cells from oxidative stress: The Role of Heme Oxygenase.

12.American Association of Pharmaceutical Scientists (AAPS) 2013: Network Analysis of the Cytoprotective Effect of CDDO-lm against Oxidant Stress in Human Umbilical Vein Endothelial Cells (HUVEC).

MANUSCRIPTS:

1. Comparison of Bioluminescence Imaging and Luminometry for Detection of Luciferase Activity in Transgenic Mice Engineered to Express Luciferase in Response to Hypoxia. (submitted)

2. Cytoprotection of human endothelial cells against menadione-induced oxidative stress by 2-Cyano-3, 12-dioxooleana-1,9-dien-28-imidazolide (CDDO-lm): a more potent cytoprotectant than caffeic acid phenethyl ester (CAPE). (In preparation)

3. Gene expression of the cytoprotective response of human endothelial cells to 2-Cyano-3, 12-dioxooleana-1,9-dien-28-imidazolide (CDDO-lm) and methyl ester (CDDO-Me). (In preparation)

4. Pharmcodynamics of H0-1 induction in mice pretreated with CDDO-Me and subjected to hemorrhagic shock. (In preparation)

7

CONCLUSION: One of the current requirements for development of drugs for treatment hemorrhagic shock is that a candidate drugs be FDA approved for some use or close to approval. CDDO-Me and CDDO-lm are in phase 3 clinic_al trials for chronic kidney disease and diabetes and as a chemopreventative for cancer development. Both CDDO-Me and CDDO-lm have been demonstrated to potently upregulate H0-1 in vitro. Synthetic oleanolic acid derivatives may become important contributors to devising polypharmacological approaches to reducing the impact of hemorrhagic shock.

REFERENCES: See presentations and manuscripts in preparation

APPENDICES: Please refer to REPORTABLE OUTCOMES section.

SUPPORTING DATA: Please refer to REPORTABLE OUTCOMES section.

8

@ Y Abstract

Abstract Number: 2128

Induction of phase II enzymes, in pa1ticularthe 32 kd stress protein heme oxygenase-1 (H0- 1 ), is cytoprotective in human endothelial cells. We previously demonstrated that caffeic acid phenethyl ester (CAPE) was cytoprotective against mcnadione­induced oxidative stress in HUVEC' largely via the induction of HO- I.• Here, we tested the cytoprotective activity of I [2-cyano· 3, 12-dioxooleana-1,9( 11}-dien-28-oyl]imidazole (C'DDO-lm). a new synthetic triterpenoid (Dr. Michael Spom, Dartmouth University) against oxidative sh·ess. Dose response studies indicated that CDDO-lm at 200 nM was more cytoprotective ag.ainst menadione toxicity than an optimal dose of C' APE (20 µM). resulting in endothelial cell survival of 80% compared to 60% for CAPE. Messenger RNA for HO-I was increased 90-fold in the presence ofCDDO-lm, while only 13-fold by CAPE. Western blot analysis of HO-I protein product indicated that by 6 h, CDDO-lm induced an 8-fold higherlevel ofHO·l while CAPE induced a 2-fold increase. The results indicate that CDDO-Jm is a much more potent cytoprotectant than CAPE, and this beneficial effect correlates well with the induction of HO· l . *Wang X. et al.. Em· J Pharrnacol. 2008 Sep 4;591(1-3):28-35.

Introduction

The induction of phase II gene products has been proposed to protect cells from oxidative stress (I). We have previously shown that heme oxygenase-1 (HO-I), a phase II enzyme, protected HUVEC from menadione (MD)·induced oxidative injury following treatment by caffei c acid phenethyl ester (CAPE). a polyphenolic antioxidant (2). To futther improve cytoprotection, we investigated 1(2-cyano-3.12-dioxooleana· l,9(11)-dien-28· oyl]imidazole (CDDO-lm), a new synthetic t1iterpenoid (from Dr. Michael Sporn, Dattmoud1 University) and potent inducer of phase II enzymes (3).

Methods

Cell culture:

HUVEC (Lifeline Cell Technology, Walket'S\'ille, MD) were cultivated on 1% gelatin-coated 48-well multiplates (Coming Incorporated, Coming, NY) in Vasculife" Medium. Only the second through fifth population doublings of cells were used.

Cytoprotective effect of a synthetic triterpenoid against oxidat ive Stress in human umbilical vein endothelial cells (HUVEC)

2Phillip Bowman, 1•2Xinyu Wang, 2.Tames Bynum, 1•2John Yang. and 'Salomon Stavchansky,

'Pharmaceutics Division, College of Pharmacy, University of Texas, Austin TX and

~-- ~f:'!.t'ij·· . ~··· ,.zjW-·

WIUU..,,_ . .. HU\ .... .;.••IU.•llh1UIUl. .. H9'1.11a!41,"'11•~,.,_,'\_,H .. 11-"'n U•.it.• llUlllU-W'•._•tJ"•A:

Methods

In vitro assay:

Cell viability was assessed at 24 hours after initiation of treatment usingA1amar Blue™ (Biosourcc International, Camaiillo, CA). CAPE and CD DO-JM were assayed for cytotoxic effects in HUVEC. Doses of C'APE and CDDO-JM cansing Jess than 90% cell viability (compare to control group) were considered toxic and not applied in cytoprotection assay. Confluent HUVEC were pretreated with either various concentrations ofCDDO·IM and CAPE or 0.1% DMSO (control) for 6 hi~. then exposed to a toxic dose of MD for ndditionel 24 hrs. Cell vinbility was assessed compared to the vehicle controls. At least three independent experiments were perfom1ed and each was done in triplicate.

HO-I induction confirmntion:

HUVEC were pretreated with 100 nM CDDO-IM and 20 µM CAPE or 0. 1 % DMSO (control) for 6 hrs (RNA) and 24 hrs (Protein). respectively. HO-I induction was confirmed using RT­PCR and western blot. For RT·PCR, die cDNA was obtained by reverse transc1iption RNA obtained directly from the h·eated cells using the Cells-to cDNA ~II kit (Applied Biosystems/Ambion, Austin, TX) and Real-rime PCR was perfom1ed on a LightCycler™ thermal c;cler (Idaho Technology. Salt Lake City, UT). H0-1 gene was normalized to the expression level of 185 for each sampJe. Relative quantification was performed with the comparative C, method. For western blot, the protein was obtained by direct lysis of the tt·eated cells and directly nm on Jnvitrogen E-page gels. which were transferred to nitrocellulose membrane using the iBlot system (Jnvitrogen Corporation, Carlsbad, C'A) Prior to HO- t antibody application, the blots were stained with SyPro Ruby blot stain (lnvitrogen) to normalize the amount of protein.

Results

120 l UlllJIJ i

~ ~ * ~ ~ n m -CA'[

Figure 1. C'ytotoxicity of CAPE in HUVEC. • : p< 0.05 versus control (0 11M CAPE). CAPE nt doses of 50. 75. ond 100 ~·M were cytotoxic and Jess thnn 40 µM were used for cytoprotection nssny.

Results

:~ llU IJ IH ~o 100 zoo 400 soo 1so 1000

nMC DDO ·IM

Figure 2. Cytotoxicity ofCDDO-IM in Hl IVEC'. •: p< 0.05 versus control (0 nM CDDO· IM). CDDO·IM at doses of750 and JOOO nM were cytotoxic and less than 500 nM were used for cytoprotcction assny.

,., , .. .. "'

" ).J[)tt MD•S MD>IO M0-20 !-ID+JO tfrit>

!'<\ICAH

Figure 3. Cytoprotection of CAPE against 70 11M MD toxicity in HU\ "EC. •: p-: 0.05 versus MD alone (MD+O µM CAPE). The cytoprotective effect of CAPE was dose dependent. CAPE at 20 1iM protected HUVEC against MD-induced toxicity (- 10% cell survival) resulting in around 600/o cell survival.

120 i ! 100 .. ., ~ 00 .. ~ 40

" : ~

~ i· .- .-.- .- -.-~ ~' §!'() _,,,rS' ,.,, .. f t:i"':<S- to)f-"' ~ ~ ... ..~ ..$' . .:;.

ti,\f<:f>f)O.f'\I

Results

:ui·~)u .. ! '">

~ 1: • I ., . " ~ j .. . " . .

Figure 5. HO-I mRNA induction in HUVEC by 6 hr treatment of 20 µM CAPE and 100 nM CDDO­IM. HO-I RNA was induced up to 90 fold by CDDO-IM compared to a 13-fold increase following CAPE treatment.

OM!-O o.,f)

Figure 6. HO· I protein expression in HUVEC by 24 hr treatment of20 µM CAPE and JOO nM CDDO-IM. H0-1 protein was induced up to 87 fo ld by CDDO-IM compared to a JO-fold increase following CAPE trratment.

Conclusions

I. Cytotoxicity profil es of CAPE and CDDO-IM were esL1blished in HUVEC. CAPE above 40 µM and CDDO-IM over 500 nM were cytotoxic.

2. The cytoprotective effect of CAPE and CDDO·IM were dose dependent The cytoprotection of CDDO-IM are much more potent than d1at of CAPE.

3. The induction of HO-I by CAPE ai1d CDDO-IM co1Telated well with d1eir cytoprotection .

4. Since CDDO-IM appears to provide improved cytoprotection against oxidative stress it is a good candidate for testing in animal models of ischemia rcpcrfusion injmy.

References

1, I. HollzdawWD, Olnkov•·Ko•toY•AT. T•l• l•y P. p,.,,1ectlona9• in•l 1JlctropNle 1nd o•id• lin

Figure 4. Cytoprotcction of CDDOelM against 70 ~tM MD toxicity ::·;~~~~~,;::;:!:.:U.:'' '°'thit~""'°' thatrtt~•tolncl11eel"$. in HUVEC. •: ~< 0.05 versus MD a Jone (MD+o nM C'DOO-lM). 2

· ~!~:~i:~vc,:~·:~; ~='~ ~;;: !c~ :~:.~~=:.:c:"n:~tvn.n The cytoprotcc11vc effect of CDDO-IM WAS dose dependent. oxy911tnHO•t. E IM' J Pt.erm•cot. 2008 Sip 4:!591(1·3):21-]5.

CDDO· IM at 100 nM protected HUVEC 11gainst MD-induced 3· ~ ~ .. ~·ci:~ :'i<:,::;~~~A~~~~;=nRM~'":~~~.?~~ toxir.:ity (- 10% cell survival) re~ulting in around 80o/o cell survival. • nd ~u:o111e. ~· potent lnd1.1«n 01 hom1 •IY9fl'l1e-1.,.., Nrt21A.Re s1gn1q. c.nc.r

which is much more potent thnn CAPE in cytoprotection. "'" '°°"""' ' '1!5("l",.._,.·

A @ Structure activity relationship of Caffeic Acid Phenethyl Ester (CAPE) and its amide derivative CAPA against oxidant stress in human endothelial cells

1•2J. Yang, 3S.M. Kerwin,1•2X. Wang, 15. Stavchansky 2J.A. Bynum a nd 2P.D. Bowman 1 Division of Pharmaceutics and 3Division of Medicinal Chemistry, College of Pharmacy, University of Texos, Austin TX 1 US Army Institute of Surgical Research, San Antonio TX

__ ___AQ_stra_f!_ Numerous reports have described the amelioration of ischemia/ reperfusion inj ury by CAPE post -injury1, and we have recently shown that induction of heme oxygenase (HMOXl} in vitro is highly correlated with this cVto-protection2• In vivo, the presence of esterases which are abundant in blood and tissues, would severely limit the effect iveness of CAPE by degr;idfn1 it to caffeic acid and phenethyl alcohol, neither of which is cyto-protective in vitro. Therefore an amide derivative of CAPE, Caffeic Add Phenethyl Amide (CAPA), was synthesized and screened for cyto-protectlon by examining its ability to induce HMOXl mRNA in human endothelial cells. CAPA was produced by a dassk W ittig reaction and was shown to be 90% pure by lH nuclear magnet ic resonance spectroscopy. CAPA was as effective as CAPE in inducing H0-1 mRNA (9-fold over vehicle control) as determined by RT-PCR. Assays utilizing CAPA have also shown that It is as effect ive as CAPE in protecting endothelial cells against menadlone induced oxidant stress. It remains to be determined if it exhibits greater st ability than CAPE in vivo.

Introduction Interruptions t o the flow of blood to an organ or t issue resu lts in ischemic injury that is exacerbated by the restoration of flow and reintroduct ion of oxygen, leading to lschemia/reperfusion {l /R) injury. Caffeic Acid Phenethyl Ester (CAPE) has been found to ameliorate l/R Injury' and protects cells from oxidant stress if'l vitro 2• This cytoprotective effect is highly correlated w ith t he expression of the heme oxygenase (decycling) 1 gene (HMOX1}2.J. As an ester, however, CAPE is subject to estaerases that readily hydrolyze CAPE in plasma and in cells, eliminating t he cytoprotective effect. We hypothesized that an amide derivative of CAPE, Caffe lc Acid Phenethyl Amide (CAPA), would be able to avoid esterase hydorlys is and b~ active for a longer period of time, while maintaining CAPE's cytoprotective ability.

,._ - Acf.,-in. - - ·~~ - - ~ •. _,.......,.

i!'

"'.'".. ~l /((lftlft/M Ot .......... -1~<1~.-..,. - .,,,_ ...... ~

~...__. o, t ~,o, •o, • ~t..._ -

··7..::- ·~ ~·~::: 0/ ~~· ·-·-.. .... __ _ •,O

~M­._

Figure 1 - Proposed mechanism of lsc-h1mla/rtperfusion lnjucy modified from Granger et at., Acta Physlo Scond Suppl, 541. 1ga6

Materia Is and~M'""e"'t~h~o~d'""s ____ _ - Human Umbilical Vein Endothelial Cells (HUVEC) used In both gene expression and cyto-protectton HHVS

-Cell RNA treated with reverse transcriptase using the method described by Ambion's "Cel ls to cDNA II"

- R~he light Cyclu 480 RT-PCR used to quantify the Induction of HMOXl cene expression, w ith the 185 ribosome gene used as a control

-Menadlone used In the cyto-protection assay as t he inducer of oxidant damage

- Cell viability in t he cyto-protection assay was assessed w ith Alamar Blue

___jynthesis

""'~) CJJ Q ~th\' l.1••- C'lt'--"1)1d!klo.W

~ -y~ l) '" THI'

·· ~ 1-'"'l •l)o\' ........... ~· i.... • • ~ .....

:xfH ' ~"*V'rot.hw~

TBDMS-CI

""""'"' 0:\1,\P

n~IF

'"" """''

• ,ro, CHP: ..... 1ilw!o00o

a!~~~J 2.11\l.l-ro-,\~'lllyl-....cl.1 111""'

~.~-/'-( 'J 0 "-''

~r·..i..-1 .... 11 , ;-,...,IJllM:r · I.. • .._,,,,._ """*

TBOMS-0, _,.... J__ TBOMS- OJvJ H

'-~~Jll-(m,....yt.diocti'·ki.a.,~t>-lk'.,..,.._,.Jlk

~;-1=~('1 """_x:i -· ·::x·~.. ,) ,__J.l "'( ) ...... , ....... • .• f

~· ---~-.i.t .............. .....-"'"'"~ i"'l c,,..,..)o...._)

·- -~ rJ ~ . ~

- I ~

Figure 2 - Stepwise synthesis of C1ffek Add Phenethyl Amide (CAPA). Fln1! compound was purif ied by flash column chromato111phy and re·crvst111ization

-----'-A~d=d~it""'io~n~a. 1 Derivatives Previous studies have shown that fl uorination of the catechol ring in CAPE Improves the stability of the compound while maintaining its cyto­protective ability A similar approach was taken here.: The followif'lg CAPA fluorinated derivatives are currently being synthesized and tested for stability and cyto-protec.tive properties.

~~CAPA

:~.r---QcAPA 1

~CAPA2

p-J-.r D CAPA 4

~_,.()CAPAS

.~CAPA3 :~~..__,() CAPA 6

Figure 3 - CAPA fluorin at~d derivatilJH, SynthHis pathways similar to CAPA

~ ~ ~ l . -~·

~ ·;; .. ~

Cyto-!:rotection

Cytoprotection against MD by CAPE and CAPA

:•111111& MO CA""·IOYM C»{.JOvM !A' ... JOuM CAM· olOwM "'

•J!.,.)l;\ID

•n .J,.M)m

. ,u,1\t\m

. ,u,..\.IMn

Fig1.1re 4 - Cyto.prottctlon of CAPE ind CAPA 1g1lnst menadiont (MO) toxicity in HUVEC

" " '

~I! J i'"

i ' ?" , ~ ~ "

__ Gene Exi;>ression __

HMOXl gene expression by CAPE and CAPA

_ .... _ OMSO ~ CAPE 20 µM

Fi&ure S - ftT ·PCR Gene expression ISS"'f quant ifying the expression of HMOXl

Conclusions --- --- --- -·There is no significant difference in cytoprotection of HUVEC against menadione induced oxidant stress between CAPE and CAPA

·There is no significant difference in the expression of the HMOXl gene between CAPE and CAPA

References - - - - -1. Ton et al., Am I Physlo Heort Clrc Phys/of. 289. H226S·H2271. 2005

2. Wang X., Euro J Phormcol., 591, 28-35, 2008

3. Katori et al., Transplant Immunology. 9 . 227-233. 2002

@ J ,

Noninvasive imaging of hemorrhage-induced global ischemia with a transgenic mouse expressing luciferase coupled to hypoxia-inducible factor (HIF1 a)

.

1P.D. Bowman, 1J.A. Bynum, 2J.A. Kiang, 10.G. Baer, 1J. Johnson, and 1M.A. Dubick,

ABSTRACT #2487

ABSTRACT Hcmorrhaeic shock leads to global ischemia. but available blood is distributed unevenly to lhc body·s orn3ns :md it is gcncr3Jly 3Cccptcd th3t blood 1s shuntro k> those organs that ma1n1am cnhcal functiom. The organs thal become most 1schcm1c and their role m hemorrhagic shock. however, have not been cktmnined. \\'c uscrt the recently dcscnbed. genct1cally engineered mouse FVB. I 29S6-Gt(ROSA)26Sortm I (Hlf I a/luc) Kael'J '" (Jackson Labora1oriesl"·h1ch ~a luc1ferasc gene fused to the rcgK>n of HIFlo. th31 binds to "°"' fl1ppel-Lntdau protein in an 011:ygcn-dcpendcnt manner generating a reporter that c-an be used to mom tor oxygen availability m mtacl t1ssuei>. Thus more light would be emitted from ischemic ussue. To dctenninc if !his mou~c could be used 10 identify organs affected b~ hemonh.ag1c shock. 40 per cent of the calculated blood \•olume was rcmov('d vta the subma:ullary \•em. and the mice were &nJCCled with lucifcnn. aneslhetizcd w11h LSOflurane. and imaged m the Xcnogen l\'IS 100 Imaging system as a func1ion of lime aOcr hemorrhage. The ventral suliace of the.c mice e'hib1tcd increased light cmiued fTOm the region of the intestine that was Jl'K'':it promment at J-6 h after hemorrhage bul still evident at 72 h. Tins technique: should allow mon" detailed t hld•ts of those 11!.SUCS most affcc1ed by hemorrha&1c shock.

INTRODUCTION Following hemorrhage. delivery of blood to organs and

tissue is compromised, resulting m hypoxia to some organs but those tissues most effected by hypoxia arc not known Safran ct al. have recently de.scribed a transgenic mouse that has been engineered to express the firefly luciferase biolumincscent repo11er fused to a region ofHIF that 1s sufficient for oxygen-dependent degradation. This mouse is designed for use m monitoring hypoxic tissues and evaluating therapeutic agents that stabilize HIFla. We asked if it could be used to monitor hypoxia induced by hemorrhage

MATERIALS AND METHODS Male mice ( 25-30 g\ of the FVB.1 29S6-

Gt(ROSA)26Soi1m l (HlFl A~uc)Kael/J strain (Catalog #006206) were obtained from the Jackson Laboratories:, Bar Harbor, ME. Mice were anesthetized with 2% isoflurane and about forty percent of the calculated blood volume \.\"3S

,v;thdrawn over a 30 second period by inserting the Medipoint Lancet (Medi point , Mineola, NY) into the subma:ocil lary H in. Submaxillary veins of sham animals were punctured with the lancet but bleeding stopped by the application of pressure from sterile gauze. For bioluminescence imaging, mice were anes1hetized with isoflurane/a1r and injected ,v; th I 00 µ I of lucifcnn (dissolved m phosphate buffered salme mto the peritoneal cavity. F1\ e min after luciferin injection, mice "'ere imaged for 1-5 minutes. Photons emitted from specific regions will be quantified using the Liv1nglmage software (Xenogen) and luciferasc activity acquired as photons emitted per second Organs (liver, lung. kidney. spleen, duodenum, jejunum, ileum, stomach, brain. salivary gland, skeletal muscle, and testes) were removed 5 min following injection of luciferin to produce a higher resolution view of the light coming from organs as a function of lucifcrasc production concomitant wtth Hlfa. induction.

1US Army Institute of Surgical Research, Fort Sam Houston, TX 78234-6315

""' - H"...._,.. Figure t. Nonim·asive imaging of the FVB.129SQ.Gt(ROSA)26Sortml (HlflAluc)Kae~J follo"ing 35% hemorrhage. The intestine appears to be most effected at 3 h post hemonhagc The throat region is probably giving off light because it became ischem1c as a result of hemorrhage via the submaxi llary vein.

Figure 2 Noninvasive imaging of the FVB.129S6-Gl(ROSA)26Sorlm I (HIF IA/luc)Kael/J following 40% hemorrhage es a function of time after hemorrhage. The same mouse was followed for 96 h before (prehcmorrhage) and following hemorrhage (posthemorrhage). At each time was mouse was anesthetized with isofluorance, injected IP with luciferin and imaged at 5 mm following lnJection The principal source of luciferasc by who1c body imaging is the intestinal regio n.

DORSAL VENTRAL

C .11 ~· ~· r-';-:·- ~=-

""1· l!lllr ..... 1111 t""·~ .: ~~.. i.5!:..:.: :.:--~:-

c-:;~.,

I . f

I" ~ -~1 ..-r

lllJ~ ~~-::.. ·~--. {~~, ~~ -:r.-·r.-.. ...

c~m. Figure 2

Prehemorrhage

3 h Post Hemorrhage

6 h Post Hemorrhage

20 h Post Hemorrhage

48 h Post Hemorrhage

7 2 h Post Hemorrhage

96 h Post Hemorrhage

I . L j .

• tJI D - ,.,_ fll!ih ---Figure 3. Quantification of light emanating from the intestinal region by luminescent imaging. The region of interest (ROI, intestine) was demarcated and applied to each time point in Figure 2. Bars represent average number of photons 'ROI at each time.

.,--c'· Figure 4. The testes as a site of HlFlaluc1ferase activity m mice. The same mouse imaged on different days. Ort the left. the testes are retracted into the body. On the right the testes have descended and express high levels of luciferase activity. This has recently been reported by Lysiak et al . Hypoxia Inducible Factor-la is Constitutively Expressed in Murine Lc:ydig Cells and Regulates JP,.. Hydroxysteroid Dehydrogenase Type I Promoter Activity. J Androl 2009 Mar­Apr;30(2): 146-56.

Figure S. Evaluation of light emanating from isolated organs of hemorrhaged :ind sham HlFlcx.~uciferase mice. Organs were immediately removed from mice following in vivo imaging and placed in wells of a 24-well multi plate and reimaged. Jn addition to high photo en1issions from components of the intestine, testes and salh o:uy gland, liver, lung. kidney, and s tomach also produced light indicating some hypoxia in these organs.

0 ..

• • • ' . __ ,

• Su mm Ary and Conclusion SHAM

[::"";;

The HJF I a 1luciferasc mouse provides a useful model for s tudy mg hypoxia associated with ischemia o f hemorrhage The high s1gnal/no1se allows detection ofHTFla activity at very low levels.

2. 2. In vivo imaging is problematic as many effected organs are not evident from in vivo imaging.

3. 3 . Even imaging of organs may not accurately reflect the degree of hypoxia as photons emitted from highl} pigmented organs may be reduced requiring homogenization of tissue prior 10 luciferase determination

Rl:fl'lffN<.:f_<;; I S.&an. M~ WY K""· F O't'Atncll, I Flif'Po, V Ciuivk:r, J. WI~ mer, R. 11.. l )qwnh..•. and \\' ( i . Ka .. 1111, Jr. M~ln\~l f<...- ncimn ''I ·<' ~ina, t•rl!IFrn•lyl h}-dr\ )'ll>i1: 11K.1l\ it . a« .. \ nw,.'l1t ••r1111 (11111 &f' .. ,...11ha1 ~imu1al~ t."l)'Lhl\'l"l"k:lin rti-"hl(1 k•u . l'n-.c N.111l Ae11d ~US A 1{K)f, ~ 10.l : \05-l IO ~ i.'. ltA•lf, N.,A. Wl.'bcr, U. 1.i,R. T.,\'lb. r R Contaj ,A r P\IJ'Ch;.i,... anJO n Wcs.1.Annlrlucit>~

mtnc: ~,. JJc \yn1tv ....... kal..,(aa<;c n!J1("1-.::r ¥Yik.'flt fN' 111 1 "'' k."\lttlJ (If 1'1'11J.mn111wn.itory Cl-.11~ '° '"'n-~K-~Jln1n'IUOl"'ll'.?00l;l7(t (1107-6.l19

""" ~~XAs

y Cytoprotection of Human Endothelial Cells from Oxidative Stress by e·

Polyphenols: the Role of Gene Expression versus Direct Antioxidant Effect · - - ~ l.2Xinyu Wang, 2James Bynum,1Salomon Stavchansky, 2Michael Dubick, 3Robert Hackman, icarl Keen, and 2Phillip Bowman ·~ _ I

1Pharmaceutics Division. College of Pharmacy. University of Texas, Austin TX: 2United State Army Institute of Surgical Research, ~:S! Abstract

Absu·acf Number: 494 l

Polyphenols have been implicated in protecting cells ag;:iinst oxidative stress. Several pol)·phenols including caffe1c acid phenethyl ester (CAPE). curcumin, rcsverauol, caffcic acid. catcd1in, and Oligonol™, a commercial source of a mixture of polyphenolics were investigated for their cytoprotcctive effects and effects on the transcriptional activity in an ;11 dim model of mcnadione-induccd o<cidative stres) in human umbilical vein endothelial cells. CAPE. curcumin. and resverarrol showed dose-dependent cytoprotec11on against menadione·induced cytoto'ticity, whereas Olii;onol™, (+}-catechin, and caffeic acid did not. Tite results of ' direct' antioxidant capacities of those compounds by 2. 7·dichlorofluorescm assay indicated that most compounds tested showed good free radical scavenging abilities C''CCept resvel'alrol. How('vcr. 'direct' antioxidant activity did nol correlate well with their cy toprotecrive effects. Gene expression analysis with whole genome microarrays and submission of s1arisu cally significant resulls 10 Ingenuity Pathway Analysis showed that a number of genes were up· or dowrHegulated by these compounds effecting common molecular networks and compound·specific molecular networks, which may account for their benefic ial effects, in particular the heat shock protein family and IL-8 sit;nahng pathway.

Introduction

Polyphenols have been repo11ed to provide beneficial effects, including anticancer, antibacterial. anti -inflammatory, and antioxidant activities. To better understand the purpot1cd protective properties of several polyphenols, including caffeic acid phencthyl ester (CAPE), curcumin (CUC), resveratrol (RES), caffeic acid (CA). catechin (CAT), and Oligonol"'. a mixture ofpolyphenols derived from lychee fruit an i11 vitro model using menadione (MD)­induced oxidative stress in human umbilical vein endothelial cells (HUVEC) was investigated [I].

Methods I n vitrfl assay:

Confluent HUVEC were pretreated with test compounds or control (0. 1% DMSO) for 6 hrs, then exposed to MD for an additional 24 hrs. N-acetyl cysteine (NAC) served as a positive control, providing complete protection against this stress. Cell ' iabili ty was assessed u sing Alamar Blue. The experiments were done in triplicate. IntracelJular production of reactive oxygen species was C\·aluatcd using fluorescent probe 5-(and-6)-chloromethyl-2',7'­dichlorodihydrofluorescein diacetate (CM-H2DCFDA).

Gene expression 2nalysis:

Total RN As from 6h-pretreated or control HUVECs were isolated and labeled for microa1Tay analysis using agilcnt whole·genome microan·ays. Microall°ay data analysis and statist ical comparison were

perfonned using BRB Array Tools (hmr/1inus nc1 mh gov/BRB­ArrayJools html). Genes were considered statistically significant with P value ·: 0.001 and FDR (false discove1y rate) value < 10%, and genes significantly altered in their expression were submitted to Ingenuity Pathway Analysis (IPA) for further pathway investigation

San Antonio. TX 78234: and 3Department of Nutrition, University of California at Davis, Davis. CA, 95616.

Methods (www ingenmtycom). IPA used Fischer 's exact test to calculate a p· value detem1ining the probability that each biological function and or disease assigned to thal data set is due to chance alone. Significant biological networks with a score greater than 30 (P < I 0-30) were merged. Genes are represented as a single node in tl1e network. The intensity of the node color indicates the degree of up- (red) or down­(green) regulation.

Results L Cytoprotection assay:

·~ 1 J:l A • 1· • (:1 -~·)Jl.l

·~ · ·~ I • ' ( M

f • ! :1 -- --I I -CU'lll t U I U S It -• COC: 11

i::. c • • • • ':L D -.. . 1 • ' N t• , . • < N

I"' . f -1 20 ~ ,. :...:._~ ~ ~ ~lllll , ; I

(llMU. I I U J ,. ~ :i :. I:. ==-~l -- • ••••• .; L

r: ' ~ : ~

~ :------- I. ff u._._1-.

... ~" ----- .. - • 4 4

::':!: ~ " -· =~I .;

Figure I; Cytoprotcc:tion assay of \·ari<H1s polyphcnol$ OGainst MD-Induced o'i1d1ll\'C stress m HUVEC. CAPE fA). CUC <B). amt RES (C) sho\\cd dosc-dcpcndcnl crtoprolcction. while CA (0). CAT (E). and Oligonol (F) showed no cytoprolcctive effect :ll 1hc non 1oxie doses teued, • : P<0.05 versus MD :alone

2. Cell-based antioxidant assay: 16fli ....

t l'U ;

L~ J i . j ; ('IO

I ~ x

,.

I> AES Control

1 CUC

-- .. r--_:..-- ~T . . _..!------ i _J .-r ... --r------ - CAPE,Ollgonol

.. .. '"" 1:0 ... TinH• fniln)

Fig11rc 2. Direc t antioxld:mt :ietMtics ofpolyphcnols in HUVEC dcmo11str;atcd by mcasurin; the alteration of iutraecllulnr rcac1h·c OX')'C,Cn species (ROS) level. After treatment with CAPE. CUC.CA. CAT. or Oliaonol. ROS wu decreased compared to thal of conlrol. while inc11b.1tton with RES aencrated ROS similar to control.

Results 3. Gene expression analysis:

Table 1: Number of genes significantlr changed by test compounds (P < 0.001, FDR 0. 1.•:IDR < 0.~4):

CAPE CUC RES CAT CA Oliaonol'Pl

total 208 1940 176 202 44 43•

UP· reguliltion l'" Jn 71 44 10' Down-regulation 76 1563 !OS 15a l8 33'

4. fngenuity Pathway Analys is (IPA):

- - ... _ .. _

- .. ,.. v

~

..:. .......

Figure J : Nccwork 01101)1i1 of CAPE. The n1ost sicnilic:int biological networks were merged with (unctions in\·oh ing c:cll cyck. DNA rcplicarion. mct:'lbolism. and cn11onical pothwnys associated wilh protein kinosc A and xcnobiotk metabolism sign:ilina.

_ .. - . - ,-

... .:...

Fi~urc 4: Network Lllltilysis of CUC. The most signific1m1biological11c1worl.:s were merged with functions invoh·ing RNA posHranscriptionnl modificiu ion. 'enc e:tprcss ion. metabolism. nnd c:anonic;il p::a1hw8)'S associated with AMPK. NRF2-mcd1;itcd oddath·c sucss rcsponic, 01~ xcnobiolit mct3bolism s1gn3ling.

Results

.. ..:: - ...

( , .,;,

... Figure S: Network o.nolysis of RES. The most stgnific:int b1ok>gical nctwOfk:s were mcracd with functions ii1"ohi11g cell cycle. DNA rcplie:1tiot1, mcbbolism, •md c1momc.al pa1hway1 associntcd with protein ldnose A and xcnobiotic mct:ibohsm sii:n11li11c.

Conclusions

I. CAPE, CUC. and RES showed dose-dependent cytoprotection against MD toxicity in HUVEC. while CA. CAT, and Oligonol did not. Cytoprotection did not co1Tclate well with antioxidant activity as detennined by the CM­H2DCFDA assay, as almost all tested polyphenols showed similar free radical scavenging activity except RES.

2. Gene expression profiling and submission of genes significantly altered in expression to IPA identified common pathways among the cytoprotective phenolics CAPE, CUC, and RES such as protein kinase A signaling for CAPE and RES and xenobiotic metabolism signaling for all three compounds. In addition, CUC induced different pathways including AMPK and NRF2-mediated oxidative stress response s ignaling.

3. CA and CAT did not induce pathways similar to the cytoprotectants. suggesting that these pathways may provide an understanding of the cytoprotection m echanism . \Vhilc Oligonol showed good free radical scavenging activity. it did not activate HUVEC transcription event significantly. which may account for its lack of protection in this HUY EC-MD model.

Reference

I. Wang X, Stavchansky S, Zhao B, Bynum JA, Kerwin SM • Bowman PD. Cytoprotection of human endothelial cells from menadione cytotoxicity by caffeic acid phenethyl ester: The role of heme oxygenase-1. E11rvpea11 Jmmwl of Pharmacology 2008: 591(1-3): 28-35.

Induction of Hypoxia Inducible Factor 1 Alpha (HIFla.) by Caffeic Acid Ph en ethyl Ester (CAPE) and Caffeic Acid Phenethyl Amide (CAPA) in Mouse Skin Fibroblasts

Ashish R astogi1•2 , J ohn Yang1•2, Xinyu Wang1.2, James Bynum• , Salomon Stavchansky2 and Phillip Bowman1

Abstract

Abstract# 5877

Esters and amides ofcaffeic acid (CA) such as CAPE and CAPA exhibit significant cytoprotcction of a variety of cell types from oxidative stress while CA docs not. A role for HIFla in cytoprotection has been described and to better understand the mechonism, we invcsrigated skin fibroblasts from the FVB. l 2986· Gt(ROSA)26Sortml (HIFI aluc)KaeL'J (HIF laLuc) inbred strain (Jackson Laboratoiy. Bar Harbor, ME). This mouse tRosaLuc) has been genetically engineered to express lucifcrase in conjunction with accumulation of H!Fla. Rosa Luc mice skin fibroblasts (MSF) were obtained from skin specimens dissociated in 0.1% collagenase for 3 hand cultivated in 10% fetal bovine senm1 in alpha Minimal Essential Medium. Passages 1-5 were used for these studies and subcult ivated into 96-well mulriplotes. CAPA was synthesized in the lab and CAPE (Cayman Labs, Ann Arbor, Ml) and deferoxamine (DEF), a well known inducer ofHIFla and an iron ehelator, were dissolv<d in DMSO. MSF cells wer< treated for 5 It with thes< dmgs and assayed for luciferase activity (Promega, Madison. WI). CA was inactive hut CAPE and CAPA were significantly better than DEF in producing a luminescent signal from MSF. These results indicate that CAPE and CAPA may directly activate HJ Fla by inhibiring prolyl hydroxylase thereby prev<nring th< degradation of nascent HIF la.

Introduction

lschcmic preconditioning has been shown to provide significant protection from a sul>scquent lethal ischcmic cvent1• HI Fla. a master regulator of the hypoxic response has been implicated in ischemic preconditioning. To develop drugs that produce a similar C)10protcctive effect we are screening generically engineered mouse cells that express luciferase when HI Fla accumulates. Deferoxaminc induces HIF la by inhibiting the activity of F«• dependent prolyl hydroxylase which is required for activation of the oxygen dependent domain of HJ Fla and was used as a positive conn·ol. CAPE has been previously suggested to inhibit HIF la prolyl hydroxylase'. We have found that CAPA and CAPE, which induced cytoprotecrion against mcnadionc-induced-oxidative stress. also induces Hf Fla and this may explain their cytoprotective effect.

Methods

I. Ctll Culture: Mouse skin fibroblasts (MSF) were obtained from Rosa Luc mice dissociated in 0.1 % collagenasc for 3 h, and cultivated in l0% fetal bovine scrum in Minimal Essential Mediwn. Passages 1-5 were used for these st11dies and were suhcultivated into 96-well multiplates.

1\ S \ rmy lnstilu1 t or Surgiul R f''tarch. Brook t Army Mtdiu l Cen tt-r, S111n Anto nio. Tuar 78234. lOivh:ion of Phnrm1ceutics. Collete of Phar macy, The l 'niven ity of Teu s :U Au~lin. 1 exas 787 12.

2. Dose and time-dependence of induction: Stock solutions of CAPA (synthesized in the lab), CAPE (Cayman Labs. Ann Arbor, Ml). and deferoxamine mesylate (Sigma Aldtich, St. Louis. MO) were prepared in DMSO. MSF cells were treated as a function of concentration and time and assayed for luciferase activity (Promega, Madison, WI). Induction of H!Fl a was estimated by luminescent intensity.

3. Cytoprotection: MSF cells were treated with 0.1% DMSO, 1.0 µMCA, CAPE, and CAPA for 5 h. Following dosing, the cells were treated with 30-50 µM menadione to induce an oxidative sn·ess and after 24 It \'iability was determined.

Results

I . Dose and time-dependence of induction: Deferoxamine \\as used as a positive conn·ol in this study and the effect of deferoxamine on accumulation of HIFla was initially studied. The MSF cells were dosed with' arious concentrations of deferoxamine for 5 hand H!Fla acrivity dete1mined. Fig. I shows that the luminescence intensity decreased with decreasing concentration of dcferoxaminc. A median concentration of 100 µM was selected for the rime course study. Fig.2 shows incubation of MS F cells with deferoxaminc at different time points. An incubation of3-5 h showed significant Jumincscence indicaring significant induction of HIF la.

Fig. 3. CA was found to be inactive but CAPE and CAPA showed luminescence as a function of concentration gradient Jn Fig.4, the MSF cells were dosed with the highest dose of2.SµM CA, CAPE, and CAPA. As earlier observed with deferoxamine, both CAPE and CAPA also showed time dependent luminescence corresponding to H!Fla induction. These results indicate that CAPE and CAPA may also induce hypoxia by activating Hlfla through inhibition ofprolyl hydroxylase.

2. Cytoprolection: CAPE and CAPA were found to be cytoprotectivc against menadione (Fig. 5).

'I U-00000 1

15" 1000000 ;

•t j f i 100000

r .) jt "~~

h ~ 11 j1 J1 h

~

Fia I MSF cells \\ere dosed wuh dc(crox11mi11e solu1ions (20-200 ~1M) ror ' h A0. 1~9 DMSO wM Uied ns n ncgatnc control. O;ita is rcpn:scn1cd as mc:m+st::indord dc\'hition. • : P<O,OS versus 0.1% DMSO :.lone.

-Is •JOOOO •

ii= ~ . n:: . 1!~ _ I ~ n 11- ~ [

i !1 II II I II t H . - ·- ;- - :- • "I j[ !I ~[ ~ j! I h

Fi&. 2· MSF cells \\ere dosed nith 100 µl\·I defcroxaminc for up 10 5 h. A 0. 1'/• DMSO was used 3S 3 ncg;,ti\'c control. D::.13 is represented 3S meatt+sl>ind:mt dc'i o>hon.. • : P O.OS ,·crsus O.J •.4 DMSO 3lonc.

j~ """"1 .

·~ J" l ft= . ' l ...... 1 . i I """i ·

i1 ~ilill 1t1i I ti), ,,,,,,,,,,,,,,, ~' =s ~ ~': ~ s~': ~ ~ ~~~~~nnn§H§

Fi&. 3: MSF cells were dosed with soh.llions of CA. CAPE. 1nd CAPA(0.10.2 .SO ~lM) for Sh. A 0.1% OMSO w:is used ;is o neg3ti\·c co1Hrol. 0313 is rcprcsc11tcd ;is mcrm+S1n.11dord dc\'li'ltion. • · P<O.OS versus 0 1% DMSO nloue.

I ·--l . fut s;.l,- . .

h·- . · . p·..... . i

i! = l b

j! =: "'"'"~!!!!!!! i~~!~;~~!i~i~!~~~~ai.aa i•••······· .. Fia. 4: MSF cells were dosed with o.1•1o DMSO (nc~e11ivc control), 100 ,,M dcrc1oxall"linc (posithc control). and 2.S µM \A. CAPE. CAPA for up 10 S h. O:it;1 is rcprcscn1ed as mcan+stondard deviation •. P" 0.0~ ,-:rsus 0.1% DMS03lonc.

140 l 120 I

·~ 1 .. a Mc11aeOM:e ..M • M•nlkl'9n• 30 IOM IJ ,_""""!Jflft ...01.'M O hhnact-50..W

.. j : i.l.I.Jl.J.A.IL t.

OMSO C::A~A

Fis. ~ : MSF cells were dosed with 1.0 pM CA, C l\PE. CAPA and cell viability mc:isurcd oifter dosma wi1h ~0-SO i1M mcnadionc A 0.1% OMSO w35 uM:d ;,s a nccat1n control. Oatai is represented as mcan-+standard dcnauon. • P 1) OS \'CfSllS 0.1% DMSO fllonc in indl'\ id11al croups.

Conclusions l. Both CAPE and CAPA induced a dose dependent increase in H!Fla in the range of0.1- 2.5·r1M and 0.5-2.5-riM, respectively. CA was found to be inactive.

2. By 2 h oftreahnent with CAPE and CAPA. a significant induction of HI Fa was observed.

3. A l.O µM dose ofCAPA and CAPE that induced significant HlFla were found to be cytoprotective at 5 h after the beginning of treahnent. CA did not show any significant cytoprotection.

4. The accumulation ofHIFla may play a role in the cytoprotective effect of these compounds. The MSF provide a simple method for screening other drngs for C)1oprotection against oxidant stress.

References I. Mllfl) CE. Jennings RB. Reimer KA (19fe6). •Preconditioning with 1schcmia:

3 delay ofkthal cell injury in 1schcmic myoc3rdium•. Circr1lottrm 14 (S): 11 24-36.

2. Choi D. Han J. Lee Y. ChDI J. Htin S. Hong S. Jeon H. Kim YM. Jung Y (200~). "Coffek acid phencthyt ester is a potent inhibitor ofHIF prol, I hydroxyl:ise: slructur.:il on3Jysis and ph:1rmxoloaici:il implicotion ". J Nutr Bioc:hcm. Sep 7. IEpub ahead or pr1111l

@ Poor Correlation between In Vivo Imaging and Production of Light by Organs in Transgenic

Abstract

The FVB.129S6-Gr(RUSA)26So1"• 111llFIA ,.,,,K.~l/J(Jackson

Labs) mouse strain (Safran M . et al. PNAS. 2006, 103(1), pp I 05-110). genetically engineered to express luciferase when the hypoxia-inducible factor la (HIF-lo.) accumulates, was used to identify organs most affected by the ischemia of hemorrhage. Lucifeiin (SOmg/ml) was administered subcutaneously by osmotic pump (Alzet) implanted 24 h prior to hemorrhage. Forty percent of the calculated blood volume was removed with a lancet through the submaxillary vein. In vivo imaging at 4 h indicated affected organs. Mice were then euthanized and portions of each organ reimaged or frozen for homogenization of tissues followed by luciferase quantification using a luminometer. The ratio of relative luminescence units/mg of protein for hemorrhage versus sham groups were 3.0, 1.8, 3.2, and 1.2 for lung, liver, kidney, and spleen, respectively. The hemorrhaged mice showed differential expression compared to sham indicating upregulation ofHIF- lcc .The luminometer method was found to be more precise than the in vivo imaging for the dete1mination of effect of hypoxia in different organs. Comparison between in vivo imaging of the whole animal , imaging of isolated organs, and luminometer readings showed a poor correlation between these methods.

Introduction

A central problem of devising treatments for hemorrhagic shock is identifying organs that are affected and determining how they respond to the global ischemia of hemo1Thage. As blood flow is redistributed following hemorrhage. some organs are more affected than others. To address the question of most affected organs we studied HIF-lcc, the master sensor of hypoxia as a guide and investigated the use of a transgenic mouse engineered to expresses the luciferase gene in tandem with H!F-lcc for identifying affected tissues. In vivo imaging that uses bioluminescence provides a non-invasive and real time method to acquire longitudinal information from the same live animal making it attractive as a high throughput technique . In vivo imaging was first used followed by imaging of isolated organs. Homogenization of organs followed by quantification of luminescence by luminometer proved to be the most sensitive and reliable technique.

Mouse Engineered to Express Luciferase in Response to Hypoxia Ashish Rastogi1·2, .James Bynum 1

, Salomon Stavchansky2 and Phillip Bowman 1

1 US Army lnstilule of Surgical Rcsurch, San Antonio, Tuas 7823, . 20ivhion of Pharm:-ce11tics, College of Pharmaty, The University ofTtxaJ at Austin , Tex:u 78 712.

Materials and Methods Results

I. Mice and Hemorrhage - The FVB.129S6-Gr(ROSA)26-So1'"''~11Fl.Lt..e1i:.,l!J (Jackson Labs, Bar Harbor, Maine) mouse strain [I} is genetically engineered to express luciferase when the hypoxia-inducible factor la (HIF-lcc) accmmtlates. Hemorrhage is accomplished by removing 40 % of the calculated blood volume with a lancet via submaxillary vein.

I. In vivo imaging for determination of organs affected by hypoxia -

2. Imaging -A 50 mg/ml solution of potassium salt ofD-Luciferin (Caliper Life Sciences. Hopkinton, MA) was prepared with phosphate buffered saline, pH 7.4. Continuous delive1y of luciferin was achieved using osmotic pumps as described by Gross et al [2).The osmotic pumps are filled with luciferin solution and implanted on the dorsal side of the mouse. Four hours after hemorrhage, mice from both groups were anesthetized and imaged using !VIS to reveal light coming from organs as a function of luciferase production concomitant with HIF-la induction. Mice were then euthanized and po1tions of each organ reimaged or frozen for in vitro luciferase quantification using a luminometer.

""''=!I ti rl v'""' 1111 ••

fig. 1: Non-in,,ash·C' imacinc of FVB. I 29S{)-Gt(ROSA 126 -!*,,-11n r1Q1~ ""'; I mice arc shown Images arc of 11 rcpn:scnt.eth c mouse from coch group. A reference in1cnsit)• scale whh units in counts or photons cmiucd is also shown. Ahhout;h biolnmincsecncc m the bcmonhagc group is higher than lhc shum. :J qu:lntit:itivc cstim::itc c~nnol be made wilh the images.

Sham Hemorrhage

II. Quantitative analysis of organs from In vivo imaging Tabk 1: Average biohnninesccncc intcnsily (ut counts) or or;;ms os obsernd from the in yj,;o imoccs of hcmotrh.a&e and sham croups arc shown. Not all ore.ans can be identified in the images and hence t~)' cannot be quantified. Kidnc) son dorsal imlges :md te stes and li\'cr 011 \'entral images were c<>sily qm111tificrl than the otlM;r organs The testes from the hcmonha;e group exhibited lc~r bioluminc.sccncc than the sham. lil·cr sho\\cd higher bioluminesccncc whereas. kklney ihowed no s1gnifican1 d1fTcrcncc between the croups. (N• J ; •p > 0.05)

Teste.s

Hemorrha2e t 65.4"' 60.2

Sh•m 359.6% 53.6

Ra1;0 IH/Sl o.s·

Ill. Imaging of Isolated organs at 4 h

3. Statistical Analysis -Levene's test was used to access the homogeneity of variance. Student's t-test was used to analyse differences between sham and hemorrhage groups in Tables I and 2. A difference of p value ..: 0.05 was considered significant.

Sham Hemorrhage

IV. In vitro analysis of homogenized organs using a luminometer

Lune Liver Kidnn Soletn

HtmorrhaPt 3.2±0.9 13.7 "'3.2 ~.4± I.I 4.1± 1.2

Sham t .U0.4 7.6"' 2.3 t.7 ± 0.3 3.4 ± 1.7

Rario !HIS) 3.0" 1.8" J.2• 1.2

T0:blc :? : A\'crocc himinclCCl\CC: (in millions relotive h1mh1cscc11ec units) or org:111s isol:11cd from hcmorrhngc n11d sh'1m group ~ nrc shown. The lumiuomctcr nnnlym shows hcrnorrhngc 1.1,ronps hove hi, llcr luminescence vnlucs th:m the sh01m group 1ndicnting h~·poxia in lhc hcmo1fhilgc gronp tN-1; •p > 0.0~)

References

Linr Kidnev

117.8 "' 62.9 54.8 ± 45.5

74.6"' 32.4 47.0±23.S

u · 1.2

Fie,.:?: Imaging of \'3rious orcans ancr removal from :a sham or hcmonh.aioc: animal. Images arc from a rcprcscntauvc mouse from each group Intcnsny of biolumincsccncc c~ibitcd from :an orsan from anim<>ls belonging to the lo'.lmc group wos Yorilblc . Not :111 org:ms exhibited bt0h1mit1csc.cnc.c in each a111mal.

Conclusions

1. The data shown in Tables I and 2 indicate a poor con-elation between organs by in vivo imaging and in vitro luminometer analysis for the HIF­lcc transgenic mouse. Luminometer data rather than in vivo imaging analysis confirm our 01iginal hypothesis that hemorrhage animals will show higher luminescence due to accumulation of HIF-lcc.

2. Some affected (hypoxic) organs did not yield significant amounts of light by in vivo imaging.

3. The luminometer was found to be more reliable than optical imaging in evaluation of hypoxic organs.

4. The transgenic mouse produces useable infonnation if organs are isolated, homogenized and HIF-lcc activity determined in a luminometer.

Safran M. ct el. Mouse model for nonin\'Hi\'C ima;:ina of HIF prolyl hydroxylasc activity: Assessment of an orol a~cnt th;it stimulotes crylhropoictin production. PNAS. 2006. 103(1). pp 105-110 Gross S. cl : I. Continuous delivery ofD,\ucircrin by impl:mtcd micro-asmolie pumps enables true rc11l .. imc biolumincsccnec im:igin;: oflucifcr:isc oclhity in ,;,·o. Mol imaging . 2007. 6(2). pp 121-.'.\0

T"W- ll..,• •11i.... ~--....i•-i!l• .. • "'1l•1W .\ahrl1IWl'lhn.\rt.1ki..,..-Mi.1.\~Wol~nJl•1fll4atw.. .. 4llaK<'•d•1m·•lllollo.r,."'"""'"M1MC'..wl .. ~llK Ott.,.fl.'v-6La"""l_,. A_.•ll.._

~ Time Course and Network Analysis of 1-[2-Cyano-3,12-dioxooleana-1,9(11)-dien-28-oyl]imidizole (CDDO-IM)

Induction of Cytoprotective Genes in Human Umbilical Vein Endothelial Cells (HUVEC) Against Oxidant Stress 1Xinyu Wang, 1James Bynum, and 1Phillip Bowman

PCOM

1Philadelphia College of Osteopathic Medicine-Georgia campus, School of Pharmacy, Suwanee, GA, 30024; 1United States Army Institute of Surgical Research, S an Antonio, TX 78234.

Abstract Results Results Results Abstrnct Number: 1597 C'DDO-IM. a synthetic trilCtl)Cnoid-dcrh•cd compound exhibits cytoprotci::th·c

activity, possibl) ,;a transcripuonal activation oflhc phase II response. To undcrst:md the ccncs rt'Spc>n.siblc :md s1cn:1lin; pathW:l)'S mYol\·cd in initfoling and dming th is effect we pcrfonncd gene uprc:uion annlysis with whole ;cnomc micro..·mnys at 0 .5. l. 3. <i :md 24 h following 1rcatmcnt with a c~ toprotccti\'C dose o'200 nM rand compared it to O. l'Y. DMSO \'chicle control. Microan"Dy d.it.1 wc!'c :malp.cd us in; BRB am~· 1001 th:it idcn1irted about I 0.000 ;cncs lh.::at were s1puficantly ahcrcd followmc CDOO-IM ttcMrnenL Submission of these genes nltercd in 1hcir expression b)· greater than two fold to Ingenuity P.1thwo.~ Analysis (I PA) indic::itcd Sc\·cr:il cnnonic:il p:i.thw:i~ s wtTc import:mtly in, ·ol\'cd in cy1oprotcc1i.vc fmmiott. Arnone lhcm. nrf2-medi<1.1cd o:cidali,·e strcn: response gene$ L."1\own to ac:ti,·atc the phase II response were .$Ol1\C of the carliut to be: upn:gubtcd. In 3ddition. CieMS for FOS :ind JUNB 1.Mt fonn the AP-1 tQnscription compln: were expressed ::11 h1eh lc,·cls at O.S ;:md I hr CDDO-IM lrti'.ltmcnt i'.IS \\Crc the enrl) eron1h rt.sponso ecnes such ns EGR!. Express ion of these genes may drive the nrf2 pathway including the induction of heme o~·gc:nasc-1 . heat shod .. protein family ONAJ (Hsp40), &Jutrunatc<) Steine li&3SC c31:ilytic subunit, and NAO(P)H:quinonc oxidorcduct.uc :ificr ~ and I. hr CDDO­IM trcaln1cnt

Introduction Tissue damage from oxidative stress. in particular from ischemic injury that occurs during a heart attack, stroke or traumatic injmy is a common occurrence that might be reduced with appropriate drng treatment. \\e previously reported that I [2-cyano-3.12-dioxooleana-1,9(11)-dien-28-oyl]imidazole (CDDO-lM), a potent phase II enzyme inducer, provided cytoprotection against mcnadionc (MD)·induced oxidant stress in human umbilical vein endothelial cells (HUVECs). To investigate the genes responsible and signaling pathways involved in initiating and driving this effect, we perfo1med gene expression and bioinfo1matic analysis with whole genome microan ays following treatment of human cells with CDDO-IM.

Materials and Methods

Cell Culture: Replicate cultures of HUVEC were cul tured as described (I) and !rented for 0.S. I. 3. 6 and 24 h with CDDO-IM (kindly supplied by Dr. Michael B. Sporn. Dartmouth University (0.2 µM) or vehicle (0.1 % DMSO). Microorroy: Total RNA was isolated and labeled for micron1ny analysis using Agilent human whole genome microarray according 10 the manufacnirer·s instructions. Dat a Ana lysis: Microarray data analysis was perfomted using BRB Anay Tools (http:l/linus.nci.nih.gov/BRB-A1,-ayTools.h1ml). Genes were dctc1mined to be statistically altered in their expression with P ., 0.001 and fa lse discovery rate (FDR) < 0.2%. and were submitted to Ingenuity Pathway Analysis (IPA) for further investigation (www.ingenuity.com). IPA maintains a large knowledge database of modeled relationships between proteins, genes. complexes. cells, ri sS\1es, dnigs. pathways, and diseases generated from published reports. IPA perfomu Ftsher·s exact test to calculate a p-value detennining the probability that each biological function and/or disease is due to random chance. The scores for networks represent the negative log of the P value. Therefore, scores of 2 or higher provide at least 99o/o confidence of not being generated by chance alone. Genes are represented as single nodes in the network. The intensity of the node color indicates the degree of up- (red) or down­(green) regulation.

Gene e>epresslon analysis

After intensity filtering. normalization, gene filtering. and class comparison between COD-IM and control groups. about 2500 genes were significantly altered in their expression at some time point during the time course study.

~; H !

11~ii-• ;;:;~~E;Jl~--~== -~

Figure 1: Dendrogram for gene up- and dowtHegulation following CDDO-JM l0.2 µM) following treatmen t Hierarchical agglomerat ive clustering "'"aS performed with Cluster and Treeview (2).

Rcfuence I ~t.n.k:-1urc-M:hvt~ rd:itionshlP' In lhc: ~~1ol'h)tcd1vc clTcc.C of ~rre.c acid phcnc11'" I c.~cr (CArE) and nuonnah:d d('l'l\'illt1..-.:.1' dfo:h °'' lk..'mC 0X)·g..,1:i~·I mdu~hflll and 1nll~'idant awv111C11, Wing X. Sla\'dumsJ,-y S. Ken\ln SU, ll(lwm:m PD E\11 J Pharmacnl 1010 Jun W.6:>5' 1-J) 16.22 2 Clusicr ana~·s~ Md d\ip~· of cmome-.,.. c.-q»~ton Jlri'flan.' F.isim MB. Spcllnian PT. Rnmnro. Boutcvt O ProcNXIAc:adSc1 USA IV>'" I>« I 9Si 2S) 14S6J-8

Ingenuity Pathway Analysis (IPA)

Genes significant altered in their expression (about 2500) were submitted to IPA for bioinforn1atic analysis. Table 1: Top up-regulated genes in early response (0.5 h. fold change> 9.9) lo CDDO-JM treatment:

s~ P.MowC" .. -s-->°R-1;\I .,.,k•nccpk'l....,,. -1.pwpA......-.,I

J-ilS f9J--~~._.,p~--'°"

Yf0~2 roo.i"Jl-•t-.Mtopowrc..s.. ·~·"'- 2 11.iRl .,.,.,.,M\1hhtpc>llH I NM-IA• nu.: lurr1~~nrf'lllhfMnh J poup A ... cll'lkr3 ll i..":\~1 A.,ia!Ofof cM-"ilri 1 .n..~,..tt~ ("'(f'U~tc'\.c .... l'llf!F-'1 l.RKi) 1o,.__..idl ._. . ... ~ ...... ~)

0o:11nw o_n 1 h 111 o ?-'It ?-." I GOll~!> lit' 2.UI 0'7 1'1 11z.1· 1 ' ""1 t • .,.lo0•1n s•'' i t '~· "' Ol> ·,""' 10:0-1 ,,~-~~ :'61Z l!t1'. "' (l't). . )90' lltn

l••,'~1.(110 1%4 JO Cl4I J\.11 : ~.!fl· 1 ~< ·2 ~% 1'.'M11.;1 r.t :1w111 11s s.z<-1 s 11.\ .1 ·9-1 l*f-llO·U l• IJ •I 1-113·?.rt • 1zc'.10M N.St0z~ D.S "n ... ,.,.. · ,,~· ..i : u N' l,)0'!0141 IO 'f..l le\' .1 "'U •I \ ll

0 11"!11

~ 1 1u;n ·1,•1 12..in -1 ,;•1·~111 .a ·u

Table 2: Top up-regulated genes in later response (6 h, fold change > 9.S) 10 CDDO-IM treatment: ~~ <"-S•k liSll I • :;., , . l .t" HSP\IA . .... ... _ ., ?Oin.,.....nn l A ' 'l_taS,'; H l-1 11,.UI 165_? m .a 1•s1 (T'l.;.A.lu DuJ~1t......oiG· ••f-•h.\,_...... ~1_eu~2 :u1f 1 91s.110 '-ct·c·2~s l~iOXI hc"'" ~tcn1MO(dc:f.1 d...,,,I ~" l_Oflll;.~ , ltl71.-' 6'•'1 .~9S4 s1.1-1.1 • n IU k "'kft..-u~ .. 7 rccev4or N'· I 0011 1~ ,11161 )G ~Y 2~7 1 ..11.H ., , .. ~ [)WAlfU OidJOlsp-i01~1&t.1•bf->l)8,-lwt l :-,,,(Oo.f.l-1$ :2111:-1~ ... .,., ~1 .,, ' 1 z.•1 O<;Gr11 ..,,..blll'< •l..: .. , ..,.,,.4.-1li •ll ..... l ~'\t~O IJ\10 ·· ·~· I Z-7 S .. Z ISSI S .U' HSP.,.I MaltfwKl.1CllDapm4oHa • " KOSlOH -1.1'., ..1-IGO 11 1 1\0 -1 '1'11 HSl'HI h.,,. ,JMKL IOSIJ>lllllOUlapt"°'~· · N'-C_C"4J.t " 1 us·1111 107c 11 11°-IO')ll HSPB I 1i~ttdtod: lll:Da proltiti l N'l_OOUJO .J (l''IJ:JnL·.,1a1 11 19·)11$ SQS'Dll " 4111 l.•l.-.-I r."IGfl)l)OO 216" 1:11 ~ 4 0%1116 1515 GTFlB pn~tnsooe,.t-r.<1~JP 1"?.(0GUl4 ·, 11·77Sl6 11 .-15 1102· .. ~~ ABllD.1 •Mri•"-.._*'i__,;i ).'4_1"''UG :•I ni.1 12.1· • ~-'I llQI "1 l!M Th~ , _ __.w.•'-'c'-(l•p!Oil•p«'-it> ~' ).\l Gfi.\'11 l'Ofl 1 'lG ~t71 lflU !.MS

ARJto\PU IUtoGTP"" ~h"•t P•'*"'l' J1..i.1_1MHOfl7?.1i1un·211s ~.•9' 1ou · , s10 11Sl'O• 11..,.11hll'X'klll:f)~!l!..Win • jl.'\I 01-1.~65 ·.1 .rnt".1 u-,· ~1n .. )6QI lhV

•:···-·""11 I I 1 1: "

Figure 2: Nrf2amediated o>eidative stress response p:uhway (P < 0 OS) nssociated with CDDO-IM time­course tre11tmen1.

Table 3: Sig11ifica111 altcred genes involved in Nrf2-mediated oxidative stress response pathway following CDDO-IM time-course treatment:

.. ~,,. ... ~., ...... )( ... -------------~--;-\ii

·~ ~,m .... , ... ,_ , , " "IWl\ .t1 " "" ' " "' 11· 1

JIN.Ill tlfl .... . .,,. .~ ...... ....,,,.""'*' .. _ ........ , ..... 1 .. _:.w;1 ·<'

"'"f • :I" ~- I>'"".: ! ·· ~ 1 · 1< , .. • , .. , ._, ! t I

h"'·" "l rt>

"' Tl"'•Vlll .... . ...... -~ ....... nlt+ll•· ... - .............. ~ ... .... _.,,

....... " "' ''" ,i ....... 1....... h ~ ..... ' ' .•• ,.,..w. .... 11. ,..,.._ •• ••'l'm \ 111'• 1 ... ..i .. ,1 • ..-... 11-.p ......

· · ·11· ...._... ... .... _._. . .. .-. ........ . ..

11,,;'H\f: , ...,. .--1 . , ·:1..: !O;...,.... ....... ,,.-.. 1--.h .... :

"~·'-' .......... ;.-fu,,,,, • .a .. .... - ...... 1 .. -·

h1\ll 1 :1,.u.. ~~·· 1 ~ ... uu ....... ~ .. ... ,.. ... r ... t_......- t t · \J\t ...... . . 11,,1 .... .. ·- \.--~· · ,... ~ Pt.orU.,.IOl""-"'s .,,,_.. t· VCl6 f'lwlll~~ .........

u .•t '""' fY',Yl'lJ

'" ... ..., "''·"' ' ... "' , ,, ~i.·: ' 'l. ll'lfl1n

l'.Vltl ...wl! ,,. ""··~ .. ' ll' \ !)':

""-_ .. ot- , ... "" ' A•'""'! : 1 .... l \l.lll<A .. : '4.1 11 !'!~ ,...t- 11• ~ tj ''''"·~ ..,. , . , . ... !

"'"

i..:· ., ... 4:f.I

'c"•:' 1.'I ••• : '""'' ! 1· ~ .• II ~ !YI I · "'' ""\\I ll" I I :~ 1 : :t :•1": !t!I L'\H

lh! 1H'I , . .. t. h K

• • • , .p:-..o !.'t 1'""'f _ .... :l•i ~ ~q : t 11 H\) .

1 ~Ill ' ~1 1

.,,-.,.~... :-· ~ :·,. . ,..,, ~ :11 •'lo(:· t'HI • ' l " I .;':,.'I -C 11• .\\~1 \I V<\:1; '1 l" U • r>tt '1 .. 1.«'l l..10 1:1& • l ,1T• U'i 1111

= ..:=.=- ~~::.

~ --- :/X.

-· A. ,'> ...::=--=.. - · ,., ,

- , , .

/

..!-=:'-

"'!:".;".:.=."

.-.... - .. .:. .. .. ~ ~~ .::-· ~

• ,..- 1 • ':' _. ·r ::t. ,:., 1

_:_cl,( ...

1..:.,: ....

""

~--= ~·~.~ .... .:

• '• •'='-,.

~·1, ... .:.~:. :" ~

• '-;,.

t

·=:~· ... ----

Figure 3: Nrf2-mcdiated oxidative stress signaling pathway activated by CDDO-IM 6h treatment ( P value = 1.9 X IO""). The red color of the node represents up-regulated genes. The green color of the node represents down-regulated genes

Conclusions

•111'1 .... JG. t• ..

;;:.:!c~~-:.~=~I ~~-~~1

; •: ~~ •: ~'. ,, ~::; • ::: • ~ :~· 1.

;:::;.::==:::::-.:=-~.... ;~~~~ :;: :.:~~:~~ :.:. ;.::.·,~: Gene expression profiling and signaling pathway analysis indicated that ce11ain ¥enes (nuclear receptor family, EGRI. FOS and JUNB) were in response to CDDO-lM treatment at early stage. which may miuate a C) toprotcct1ve response of CD DO-IM in a later stage Cc11ain genes responded to CDDO-IM stimulation in a later stage including heat shock protein family and phase ll enzymes, which may account for CDDDO-JM cytoprotecl1\ c effect

'" ' " .1! ...... "'~'"' un ..... ..... r ... tJ,-i •

''" """ l · .. \i <.-~

' "'"'"

~-~ llf• !l~p.··~ • ·\.! Cler! · " '.• o· t"•ll6• ' .: J H " .•111 " .1• ,..,. "-•Dl' "'I ........... -- .. U ll>• ! ..._(/(' • ' 11•., ' \~1· • ,,,, •' I • ~

=3§a~~;!~~No~ .... 1~r:~ : : }1~:::rn1:~ :~ ••..,-.-:~,... ..,. .. ,,. __ ,_t .. - k - ... • ••., -u_l!l!\a• ' • \ I ':t-'t' ~, .•-.lal

·:·~~:;-: .. ~:::· ~:·;_.~·~· : _: ~: :-,' ,'~- : :~~;: ~~.4 : ; :.~ 3 .

~7:~~:~~~,~=r ·~~:~~ ::;:~~:;::::'.?:~ti~~[m ~;:_t~~,::.·,~e"~ ... m-~~ffi__,,~· .. ; _: :~~ ·1 :; ·,~:; '.~ :;~~ : -,t~~

Genes involved in Nrf2 mediated cy toprotectivc pathway include both early response ones and later response ones This 111dicates that Nrf2 mediated cytoprotective response plays an important role in CDDO-IM cytoprotec.tion.

@.' -~ . '.·~ T ~ · z. Determination of the Minimum Exposure Time for Effecting Cytoprotection in Human Umbilical Vein

Endothelial Cells (HUVEC) for Caffeic Acid Phenylethyl Ester 'CAPE) and Amide (CAPA)

J. Yang1, P.O. Bowman2, and S. Stavchansky1 ~~ -~ 1Division of Phormoceutics, College of Pharmacy, The University of Texas, Austin TX 1 US Army Institute of Surgical Research, San Antonio TX

bstr71r.:;~

We ha"e pre\ iously shown that CAPE and CAPA protec1 HUVEC from oxidant stress.1 This activity was co1Tclated with the production of heme <>x)'gcnasc-1 (HO-I). The objccti,·e of this study was to derenninc the minimum exposure time ncccssat) to pn1vidc cytoprotection and HO· I produc1ion.

MUVEC were exposed to cytoprotecti \ e doses of CAPE and CAPA for 0.5.1.2.J.4 and 6 hrs. The compounds were removed al the end of each time period and pl.aced in fresh media. Levels of HO-I \YCfC then evaluated by gel elech-ophorcsis and immunoblotting. and C) toprotection against H202 toxicity was measured. CAPE and CAPA both showed a significant increase in H0-1 expression over vehicle control within 30 min of exposure. HO- I levels for both CAPE and CAPA peaked \\'ilh a 4 h incuba1ion time wilh no significant differences beyond 4 houn. Significant cytopro1ccthc effects against h)drogen peroxide for bo1h CAPE and CAPA were found with as little as a I h exposure time.

CAPE and CAPA both s howed significant effects on HUVEC HO-I e....:prcssion and cytoprotection with only a brief exposure. h appears 1ha1 the compounds do not need 10 be continuall) exposed to the cells in order for 1he beneficial properties to be e~pr'CSSed. Whc1he1· 1his is due 10

cell loading of the drug or that minimum exposure to drug higgers a switch that leads to the effect is under investigation. Funding prnvidcd by the US Anny Medical and Materiel Command

I Y.-.J ~~-tH. Vl~X."--Prl.J.'.<no"'SM.St..,.."'-l,,.t .. ~.,._•u.-.,fuCr"-.-' ~ll."l~(l0 \P\1 ft_....,.c4 4cn · lll1' .. ~vMe(w•Wlf't#'ltcll\'Uft( ., .. c•lt'et ~4 pbcllcllil)1 c.Lto lt".\f'l) ' "" ,l/,...,_- ~"' :ZOi i'> lfll4)·SOtl.t

.., rOC';U'

fnlenuptions 10 the flow of blood to an organ or tissue result in ischemic injuf) 1hat is exacerbated by the restoration of flow and reintroduction of ox}gen. leading to lschemia/r~rfusion (I.'R) injury. Caffeic Acid Phenelhyl Ester (CAPE) has been found lo ameliorate l/R injury and protects cells from oxidant stress in vitro.2 This cytoprolccth e effect is highly correlarcd with the ac1ivity of the Heme Oxygcnase-1 (H0-1) enzyme. It has been shown that CAPE is rapidly decomposed and exhibits a sho1t half life in both plasma and in circulation.' Caffeic Acid Phcncth\ I Amide (CAPA) was synthesized to improve the stability and ac.ti\ iry propcnies of CAPE.

CAPE

HO , ~ ~O~ HOx:r

CAPA

~'~ HON ' "Aa. \. ~'l._...¥s.a--m 9'..,. .. -:\l ~•U.--. .. cir. 1cilu fh u4~-l ·-t<'~) Md~n,...._...1• \f't\.-...-*' .. .-1_..._~d.,_..,.. .. ,.~. • '--............, .A.<~H M~tlC lrNMDio.l<llH)O,.._.l

1 \\ '11$ .'<. r lf'Cl.!-!Mrucf1J.\. <'lll 'l\MJll:t'c4ioo<11~,ofun.1 ac:lflplt,;11<Mi>·1~.wa..J11t ''""'"°'..,~Ii 0ll('ION(C<Jlkn\'t0 k>liO'l'i ... tMl'•-IS!ll""tnlt!lfl~ r,i • ~4" ,_,-.f>l-~f>">. /#~l0(5 ~ . .!1 1.A

!v'laterials and Metl}ods

• CAPE was obtained from Cayman Chemical (Ann Arbor. Ml), and CAPA was synthesized previously in our lahorotory HUVEC \\Cre obtained from Lifeline Technologies (Walkersville, MD) and grown to connucnce at :\7°(' in hun11dified atmosphere with5%C0 2.

• HUVEC were treated with 5 µg 'ml of CAPE ond CAPA The compounds were removed and !he cells washed wi th PBS al JO minu1es, 1 hr, 2 hrs, .l hrs, 4 hrs • and 6 hrs post treatment. AOer removal of the compounds the cells were mcuhatcd with fresh media. At the 6 hour time point, l-IUVEC' were lysed ond o prote in

polyacrylamide gel electrophoresis and western blots prepared. H0.1 was quantified using rabbit anti HO· l and mouse <inti-~

actin. using Licor Odyssey S} stem HO-I levels normalized against P·Actin .

• In the cytoprotcction assay, HUVEC were treated similarly as described above. At 6 hours. media \\dS removed and replaced by MCDB 131 salts containing hydrogen peroxide for one hour Hydrogen peroxide solution v.as then removed and fresh media

reint roduced to the cells. Viability was assessed 18 hours followmg this using the Cell Titer Blue assay.

Objectives

I. To detennine the relationship between e.'(posure t ime of CAPE

and CAPA to HUVEC and cytoprotective activity.

2. To detennine the amount of CAPE and CAPA exposure time

necessary to significan1ly induce HO-I in HUVEC.

Results

Tcmpen'tturc CAPE t 112 (hours) CAPA t 112 (hours}

4 "C 5.36 N1A

2s·c 1.39 63

J7 ·c 0.30 14.2

60 ·c NIA 0.92

Table I - Plasma stability of CAPE and C APA. Male Sprague-Dawley nt plasma was spiked wllh CAPE anct CAPA to final concentrations of 100 µg/ml. Dccompositton obscr"ed al 3 1emperarures per compound for a minimum oD half lives. Concentrations detcnnincd. b\ HPLC-U\' at :'20 nm.

.,._J.M-rls--m.sar...--i.·s.~ tlC•D'nc~"'"**'IA111J• (t." U'\1ilbl ,.._ ,_ .~c:~~ 2ou

JOO

- 90

~ .. e " 9. a (iO

;; ~o

!O ... -~ ~ JO .. • 20

JO • C\I .. : CAl'A llWZ

figure- 1 - Cytoprotcction of HUVEC. CAPE and CAPA <20 uM) incubated for 6 hour.; pnorto a 1 hour exposure to hydrogen perox ide (2 mJ\ .. f ). Cell Titer Blue a~'..iy for quantification of cell v1abihC).1

1 ,-. ... 1,~<i\. 9\-.:c.o--Pf.:J.;_,.,. -,\L:a..~~s . • •• .,,......._ • .,.utr.:w.'""' ,.....1 ... (CIJ'I'.),._._......,.,._.,. c......,.-"'"'~.,-ctlf~.-•cdJcic-.1 ... "'~ .-n1<" U'£) 6¥7' '• • l .,.._•llll\O.l&cU'HJl_.

Fl tu rt 2 - HO-I 1nduc.11on of CAPE and C APA as a function of exposure ltmc in HU\ 'EC.

~ ,, !• ,,. .. _, g:?

)Jom·•t'\ pG~Ul"f

• CAPE • { ' /\I',\

Fi iture 3 - HO- I induction ofC/\PE ;ind CAPA as a function of e,.pasurc time in MUVEC. Values normahzcd 3gainst (\-Actin q

I

I "" 100 e 90

! 80 0 70 "' ~ 60 'Q 50 I

~ 40 II II & ;o '' .I l l II II CAPE

•CAPA

~ 20 . "$. 10

0 6

Hours of exposure

Figure 4 -Cytoprolcclion ofHUVEC. CAPE and C'APA (5 µg.'ml) incubated pnor to a I hour e \posurc to hydrogen peroxide (2 mM).

_ 120

e ~ 100

0 "' 80 ~ 'o 60

~ .i:: 40 :a ·~ 20 ot

0 1.5

mM 11202 25

CAPE

•CAPA • H202 only

Figure 5- Cytoprotection ofHUVEC. C \ PE and CAPA (5 ~1gfml) incubat(tl for I hour prtor to a I hour exposure 10 hydrogen peroxide.

Conclusio

I. A hour to both CAPE and CAPA was sufficient to effect significantly increase HO· I and to affording significant cytoprotection against oxidative stress 6 how·s later in HUVEC.

2. CAPE and CAPA do not need to be continually exposed to HUVEC to induce protective effects.

3. Achievement of a cytoprotective effect in vivo may only require achievement of a dose necessary to induce HO· l and may not requi red repeated dosing .

@ Comparison of atmospheric oxygen versus physiological levels on cytotoxicity of menadione

and cytoprotection by antioxidants in human endothelial cells

James Bynum1•2, Ashish Rastogil.2, Salomon Stavchansky2 and Phillip Bowman1

'US Army Institute or SurgiuJ Rrsurch, Brookf' Army Medicll l Cf'n ter, San Antonio. Ttx11 78234i 1Colle;e or Ph:u-macy, Th" Uninr1ity ofTtus at Austin. Tu as 78712;

Abstract

Abstract# 1120

The goal of m 1•i11v screens of potential cytoprotccrivc agents is to identify drugs that will be beneficial m vil'C>. HoY.ever. in ''itm screens arc usually pe1fo1111ed on cells cultured in atmospheric oxygen (AlmO,; 20.9%), which is much higher than the levels found in !issues in vi\o (3-5%). The aim of this study was to evaluate the cytotoxicity and cytoprotective effects of an oxidant generating injurant (50. 60. and 70µM menadione; MD\ and several antioxidants on hmnan umbilical '°in endothelial cells (HUVEC). Dose response studies demonstrated that 50 µM MD reduced viability by 80% at ArmO,. but only by 40"/o at 3% o,. Antioxidants (N·acetyl cysteinc, ganuna-glutamyl cysteine, an<l glutath.ione) were more effective cytoprotectants at 3% Oi than Atm02• and lower doses (5 and 50~1M) were 1nore effective at 3% 0 2 than at 20.9% AtmO, However. all three compounds proved to be cytoprotective to HUVEC against MD-induced oxidant s!Tess at their highest dose of 500µM. These preliminary results may suggest that the relatively high amount of oxygen in room air is a substrate for generating oxidants, and screens for cytoprotectivc antioxidants might be more predictive of in vivo perfo1mance if done at more physiologic levels.

Introduction

While most cell cultw·e is perfom1ed at 20.9% oxygen. several studies h8\ e indicated that cells perfom1 better at lower. more physiological levels of oxygen1·'. Low levels ofreacriYe oxygen species (ROS) can function as redox·acthe signaling messengers. whereas high levels of ROS induce cellular damage. Menadione generates ROS through redox cycling, and high concentrations trigger cell death' . N-Acetyl cysteine (NAC): Glutathione (ga1runa-glutamyl-cysteinyl-glycine; GSH). and Gamma-glutamyl cysteine (GGC). GS H and NAC are popular antioxidants kn0\\11 for their ability to minimize oxidative stress and the downstream negative effects thought to be associaled with oxidative stress. GSH is largely known to minimize the lipid peroxidation of cellular membranes and other such targets that is known to occur with oxidative sh·ess'. NAC is a by-product of GSH and is popular due to its cysteine residues and the role it has on glutathione ntaintenancc and metabolism. GGC is a precursor of GSH and is used by glutathione sythetase to fo1111 GSH in cells.

In this snldy. we aim to e\aluate the cytotoxicity and C)10protective effects of an oxidant generating i1\jt1rant (SO, 60, and 70µM menadione: MD) and antioxidants on hwnan wnbilical vein endothelial cells (HUVEC) at oxygen concentraions of 3% and 20.9%. respectively.

Methods

I. Cell Culture: Pooled HUVEC (Lonza, Walkerville MD) were culthated on 75-cm' culture flasks (Coming Incorporated. Coming. NY, USA) in Medium 131 (Life Technologies, Carlsbad, CA) supplemented with 5% fetal calf serum. penicillin (I 00 units'ml). sh·eptomycin (100 units/ ml), and Fungizone (0.25 µg:ml) and endothelial supplements supplied by ATCC. Stock culhires were cultivated at 37°C in a hwnidified atmosphere of 3% oxygen and 5% C02 with medium changes every 2 days until confluent. Prior to an experiment, HUVEC were subcultivated with T1ypsill'EDTAonto Costar® 96 well multiplates (Coming Incorporated. Coming. NY, USA) and used when confluent. 2. Oxygen Measurement: Forrna incubator equipped with Fyrite Analyzer Kit was used to non-invasively monitor oxygen levels. A Sensor Dish Reader (SOR) were used within a contTolled oxygen environment provided by microprocessor controlled chambers (Coy, Laboratories. Grass Lake, Ml). 3. Dose Response Studies: Stock solutions of NAC (Sigma), GSH (Sigma). and GGC (United Peptides) were prepared in media. HUVEC cells were concml"Cntly dosed with 0, 5. 50, and 500 µM of antioxidants and with 50-70 µM of menadione for 24 h. 4. Cytoprotection: Followmg the duration of trearmenl cell titre blue was added to the cells and cell viability was detennined as the function of fluorescence at Ex 560 nm and Em 590 nm. 5. Western blotting: Following appropriate culturing. cells were lysed in buffer and run on an 8% polyac1ylamide gel (ePAGE) and transfmcd to a nih·ocellulose membrane (iBlot: Life Technologies). Primary HIF- IA antibody was obtained from Novus (Littleton. CO) and secondaiy antibody was an from Li-Cor, Lincoln. Ne) and blots were scanned on an Odyssey. Westcm blots were quantified by lmageJ.

Results

Comparison of cells grown under 3°0 0 2 versus those grown at Ann 0 2.

demonstrated that the fonner were more effective against the menadione induced cytoxicity even in the absence of any anti-oxidants (Fig. I). Dose response studies demonsh-atcd that 50 µM MD reduced viability by 80% at ArmO,. but only by 40% at 3% O,. This data suggests that the high oxygen levels in atmosphe1·c may actually be causing oxidative injury to the cells and are thus cytotoxic.

Antioxidants (N-acetyl cysteine, gamma-glutamyl cyste ine. and glutathionc} were more effective cytoprotcctants at 3%, 0 2 than Atm02,

and lower doses (5 and 50µM) were more effective at 3% 0 1 than AtmO, (Figs 2 and 3) However, all three compounds proved to be cytoprotective to HUVEC against MO-induced oxidant sh·ess at their highest dose of 500µM . (Fig.4 ).

Additionally, western blot analysis revealed higher intensity ofHIF-1 alpha induction in cells grown at 3% 0 1 than those grown in 20.9°/o 0 2 (Fig. 5).

IX)O l>boeb1m...-r Th.: l'f'U\l\ltt« 1ll'a'~l\1Wi 1..i.>nh.-:d lu\:_• t arc th.! rn 11c ni .• "'\\°S flflhc 1U1.l1,v and arc 1k'1 k1 he ron.~lrucd •~ 1•ffte11I 1ir:t~ l'\.-fla:tinJ ~ \io;. a 1o 1•h~ ~n1111C1111•rth.!Am~ fir thclJ1..'(V11n ... ,11,•fDd.:n...:.

0 uM Compounds at SO uM Mr:nndione

I • f. .. I f .. I

Fig I: Comp:1rison of cells grom1 under ~% 0: \'Crsus those ttr0\\11 :11 Alm O~ A .SO µM MD reduced , ·inbilitr by KO-!. a t AtmO:. bul on ly b)' ~o,.. at 3% O; fn• 3J

~O uM Compounds at 50 uM Mt ni:.dionr:

..

Hlllll I Fig. 3: Antio.1'idanls : G:imma-Gh1t.am)·l Cy11cinc (GGC). Gl111a1hionc (GSHl, 3nd N·Acelyl C)•steine (NAC) were more ell"cc.lin: cytoprotee1:1n1s at:\% O; than AtmO: (n•l)

t J ~o

h -· ii ,,.,, 1'

f ,..., j 1000_0 .. I _ ............

) "02 20.PKOI

Fi.:: . .S: Jn1cnsi1y or HIF·l alphn hi duct ion was obscrYcd hi gher in cells ~ro\\n in'"• o~·y;:cn .os comporcd 10 lhosc arown 1n 20.9%0:\.)'GCn

f " ! M • ~ .. f M

S uM Compounds at SO ul\1 Menadiont.

I I I • I I Fig. 2: Antiox1d::uits : G:imm~ -Glu1 3n1~· I Cystcine (GGCl.Glu1~1hio11e (GSH). and N·Acetyl Cystcinc (NAC) were more cN'ccth·c C)'loproteclanlJ al Jo/. 0! than Atm01 (n:-3)

500 uM Contpounds at 50 uM Mtnadionf!

Hllllll Fii::. 4: AH three Antioxidants : Giimmo:i -Glutomyl Cyslcine (GGC), Glut:ithionc (GSH), :md N·Acc1yl C)·stcinc (NAC') were cffcc1i\'C c~oprotcclan11111both3o/. 0 1 1h:1nAlmO! (n•3)

Conclusions

I. Physiological levels of oxygen. that is, 3~~ Oz was found to be more cytoprotective than Arm o, against tl1e menadione induced cytotoxicity.

2. Antioxidants were more effective C)10protectants at 3o/o 0 1 than Arm o,

3. The data suggests that the relatively high amount of oxygen in room air may be a substrate for generating

References

oxidants. and screens for cytoprotective antioxidants might be Jn()re predictive of in vivo performance if done at more physiologic levels.

I 1'11ckcrl., Fuel .. K ll'w ,,. ~;_t."11 C(>lW.:CUh':!llr-11 CllC•d• lhc ur ..... f'11l of cullun.-d hum;rn dipJojd ttll,. Na1urr l'Jr,267 :4~3-425.

2 ('l11.i1Q. Fi!<ehl.'!' /\, lk::i~an ID. '1.111 IJ, Ancs J\N. 0 kbt1 c l>Ni\dam;ipc aOO \<:l1'.~'IK:Cl•fhu1nnndt,.loiJ fibrflhl.nt cell~ P'n.•c1..'0li•'l"~ M thc Nn1lflnal Ac:11.km1 1•fScicnccs <'>f lhe l hul\.-d Si;itdi 1•( An~-n:.11 1995;92·4_l37-1~1

"l Sa1'0 11. 11anvt!ll'l•d Ar. M1"'C" RE. lbc cfrcci vflcn· (' ·t1:ft '°"'iNt"" 1he tn \ ltl\>-n.1•h1:ati.-c hfc loJ'l\n oflunan J 1('tllicl fibubla~ t'\:ll' a"'1 lh.:• tnmsli.'l'llloJd do.."m.111 "' r ~n~.,'UI Cell R(:je,,111:b 1995 217 272-2~-

4 .\lkuri KR, l k:!71.,1b.:rt 1.A. NicnM AK, ('1•wa11 J , lk-T1.cnha-.1 LA ln~.tnce \•(cultW"!np: pmMI) lyi1irl« ) 1D: at J'h:-,\,1ko111.:al f ' 'l!Cll J,c\'Cl$ Pn'CN1tl M11J ~1 llS A. :!007;HM 4~7. ~55.!

5 \ll'lninclk• S. Smllf'l..'1' F, 'Krl1•lica A, (;Clkl~lcm J. Ml'J,,, S, ('1uurisi J. O .. :i-.. cn ~cn~lli\ 11y ~-.-crly hm1ti the n::rlic11ti1...: lif1;!1A f'lll c>f munr11: ribl\1bla.;1~. Na1un:: l'cll IH<>"-'Cr · .:!OO~:ll ( S), 741 · 747 f1 l.111'4'G, Ko:,iJaralh J, X hl'll..""'1..'r JM , Chcu"-ld NS. \ J1-.kJ1 lkx-t.. Tl~ ~hum.1Cki.Y t•r. M1.:11aJ;..'tlll tnf1:J'l-'11 ~"<:11dc-.lth11'w\1u:h ROS -<k.l"'dcnt l'IX\::h.1m~n.s in ~* ift·~ Pi\Rl' ~df\111i..•n • 1lhoul

n.qwrin~ 9P''f*'Sri r-n.'C RaJtc llll"I Med .20101~ 15.~9(12) 1925-36 1. K~1tk ('. ~1••u:~· I>. The a1ll• Mt.11M n•lct.lfJlu1.01th1,ll'IC ••Id N-«:d~·k.>"C•l'll! >uJ1'k:mcn1~ anJ c·~·md1.1.:cd C" id.'ll• c i'1n..-..s J lfll ~ Sp.:>ns Nu1r ~105 1),,:c 9-~;.:vi; •.• u do,\1

10 llM.l:'iSO·rU -2-2-l\.

J>harmacokinctic Profiles of Caffcic Acid Phencthyl Amide (CAPA) and Caffcic Acid Phcncthyl E~t,. .. (CA PE) i'l :v1 ale Sprague-0"'' Icy Rats

1J. Yang, 1•2J. Bynum, 2P.D. Bowman, and 1S. Stavchansky 1Division of Pharmaceutics, College of Pharmacy, The University of Texas, Aust in TX 2US Army Institute of Surgical Research, San Antonio TX

Abstract

Purpose The phannarokinct ic profile o f CAP,\ was invest1ga1ed compared. to CAPE in male Sprague-Dawley rats wnh the pu rpose of detcrmimng whci:hcr CA P;\_ 1n amide dcm at wc of C \PE. resulted m prok>ngallOn of the chmmat1on half-life from blood plasma. CAPA and CAPE have been shown to exhibit S•&mficant C)'toprotcctivc properties i11 1·/.'f (), CAPE has also been prev10usly found 10 be s ignificantly proltctn•e agautSI ischemia lrtpcrfuSKln lnJUty lll 111n.

Mcthods Male Spra1ue-Dawley rats were admm1s1ert'd CAPA al S. 10 an<1 20 mg/kg doses ~ nd ('APE at 20 m~kg (n• !i) via intravenous bolus through a surgica lly implanted JUglilar ,·cin catheter. Blood samples were collected at 8 t1mc pomts up to S hours for C.'\PA and ;l hours for CAPE. Compounds were cx1raa«1 w nh C1hyl aceme, concentrated with a rotary e\•aporator. reconst11utcd with methanol and quant11a11vely dctcnnincd using a vahdated LCMS method "ith electrospra>· io nization. Separauon was perfonncd with a Phenomencx MAX· RP l l 50x2.00 mm, 5 µ m) column. Mass sp«trometl) analr is was pcrfonncd with an Agilent 1100 smes s m&k quadrupole M SO with multunc.xk spray chamber run in ESl-ncgative mode. Plasma concentra t1on time data or CAPA and CA Pe were anal)'l.ed by WmNonhn Profe!.s1ona l (Phars1ghl). Phannacokinetic parameter'\ were anal y.ecd by non­compartmental anal}SIS <NCA) as well as through non-linear fi t 10 a

b1cxponcnua l cquafon. Rr-sults There was a stat istical S•&nificant difference found in the climinarion half-1ife txr.vccn CAPA and CAPE at 20 m~ (255. I minutes 1·i:.- ~11\ 92.J nunutcs. P < 005). Clearance was found to be 156.1, 102.6 a nd 45. I mVmm. and \'olumcord1stributton was found 10 be 52.4. J 9.2 and 17.81 for the ~. 10 and 20 mg/kg CAP..\ dose groups from NCA . Phannacokinetic puamcte~ were ;m 3lyzcd by non-a>mpartmen1al analysis (NCA) as well a, !hrough non-linear fi t 10 a r.vo C'om partmen1 model. .. Cont'lusioas lnlra\'CflOUS bolus adm1ms1ranon o f 20 mg: Kg o f CAPA. Che arnKte denvatwe of CAPE. to male Sprague Dawley rats resulted in appro.1(1mately a 170% mcrcasc in the elimination half. life when compared to CAPE. (' °'PA' s clearance and volurnc of distnbuuon WCTC dose dl'PCndent suggesting non-hnQr phannacokinet ia.

!_l}!ro~uc'on

CAPE, a natural plant product and component of honeybee propohs has been found 10 exhibit a large varieiy of beneficial effeclS such a· anti-inflammatory. anll-canccr. anll·v1ral and 1mmunomoclulatory act1\'1lles. It was pf'C\'1ously repon ed that C \ PE is cytopro1ec1we agains l mcnad1one induced oxldative stress 111 \'illn. and tlu s effect has been correlated 10 the ability o f CAPE to induce hemeoll.ygcnat e- 1 (HO- I) rather lhan lo direct antiollida nt aaivit). It was found however that CAPE is hydrolyzed rapidly ;,, utro and after in1ravenous administraoo n of 5. 10. and 20 mg/kg to catheteriu:d male Sprague Daw ley rats, CAPE ex h1b11ed a very rapid elimination. Efforts lo improve the m wm and 1'11 H \ 'O stability of CAPE led to the ~ynthec;is of Ct\PA. T he stmcturcs of CAPA and Cr\ PE arc !lhown in Figure I. This compound was Jhown lo be :1s ~1opro1cc1\\·c :1s CAPE :1g:1inst hydroscn pcro-,:idc induced Oll. idati1·e st~s m human umb1liC"al ''em cndothehal cells (HUVEC1. CAPA has also been shown to have an11-0:udant and radica l scavenging aaiv1t 1es. T he elimination half. life of CAPA was also found to be s1gnificanlly longer than CAPE m male Spr01gue Dawley ra1 plas ma, as CAPA exhibited a half· hfeof 10 hours at ~7 "C compared IO O. l :l hours for CAPE

The ob; t C11ve of the prcsenl study was to develop and vahdatc a LCl\ IS method w uh electrospray tonizatlon for 1hc quantitative dctcnninatton of CA PA and CAPE followmg e\ lr.Jctio n from rat plasma. This method was then used 10 conduct an exploratory study of the phannacokmeuc profiles of C \ PA following intra,•enous bolus admin1$1ra11on of 5, 10 and 20 mg/k& dose; to mare Sprague Dawfcy rats and compared 10 CAPE admin1s1crtd a t 20 mi'kg.

Materials and Methods

•CAPE and /f'an.f-resveratrol (internal standard) were commercially available from Cayman Chemical (Ann Arbor, MI), and CAPA was synthesized previously in our labo ratories

•Agilent 1100 series single quadrupole MSD was used for the quantitative determination of CAPE. CAPA and resveratrol

•Separat ion was achieved using Phenomenex MAX·RP ( I SOx.2 00 mm, S µm) column, water acetomtnle mobile phase in gradient elution

•Jugular vein catheterized male Sprague-Dawley rats were obtained from Charles R1,·er Laboratories

•CAPE and CAPA were administered v ia intravenous bolus via jugular vein catheter. Injection solution comprised of 45% propylene glycol. 40'/o sterile saline and 15% ethanol

•Pharmocokinetic parameters calculated using WinNonlin in both model independent and model dependent (2 compartment) analysis

CAPE

HO~O ~ I ~ ~ I 0 ~

HO .df>

CAPA

HO~O ('~ I ~ ~~ HO #

Figure 1 - Structures o f CAPE and CAPA

Results

, .. r ~ ~ _]~

·~ !" . , - . A ,, ... )

'] ~ I A ,..,.,

f!lmr£.L- l.CMS c hromatograms or A) CAPA 9) CAPE and C} Resveratrol

.. ... l" • --· ---J.. --- !ff

[ -. "' NI UO

'"t• .. • - --u-t .. '"[" "

t __. U t .. '"ti f)JJ ... _l_ O I N

"'

-:~

j :_

J ., ~--,,. - :ft

I \'~' !SO l$~

'"l' j'"' ~: ~ ___ _._LL. ·~---·--t:__ ldJ '1"J

,, .. L . . ,,~,. " ... ,,. ·: ._J _ _.i......_,._i...J,L.l i.,._

M UO f9'tl :$0 .U O

Figurt J - Full mass scan spectra of .\} CAPA al 70V fragmcnlor. B) CAPE al 10\I fragmenlOr, C) Rcweratrol at 70V f~gmentor. 0 ) C .\PA al 2ROV fragmcnlor. E} CAPE at 280\ fragrnentor. F) Rcsvcratrol at 280V fragmentor

Nomiml °""'~"' lnlr...d.)• Coftc;aatralioti ~m1rat1on prt"tlA Oll

tn1 1ml) ( 11g1ml!SD) t•.RSI))

lnlt;'l'-da~ """"". r1ccUton ,•. donlton.) 1•;.RSDl

CAH'I

20 1102± I SI 611- KX6 10 74 ,,. soo 510 4:!: lll J 196 -2 02 lB 2 1\3

2000 1991 : 61 1.98 - 420 H~ 0 47

CINt

20 2() :t7:i: 0 1(9 I 149 - 6 64 ••• IR'

'"" 4•Xl1 :i: l'JJ 245 - 52<• 4 ., 1 94

2000 1%S: 51 2 42 - 295 27' 116

Tabls I- Assay validation parameters

- 1()0(\ ' 1 i ,I r ! iii . • i 200 iii •• ~ ! .!! ..

10

. 00

• 200

l

• JOO

Time (ialauln)

• CAPE ?0 res kc • CAPA !i)tnJ lr.;i

• 'l<) ~00

~- Plasma conccntrauon·tuneprofiles for C.\P\ and CAPE

Ml(\

! a • i i .., i i >(

a 20

' . 100 200 JOO 400 '°' 000

ThM(Ulbmt"~

~ - Pla .. ma conccntra11on-ume profile.-. for C:\ PA(20. 10 and 5 mg/1'g)

,., r[ .!O•~ i..-

( 1".(i 1SD)

Jt.'("A

L 1 ~ (n1u1) n.~ a 19.~

n 1 1m11m1111 11~5!5<>.S

\ d f l ) l)_t :;: S.I

Al lC",(11c•nw..1) S9-' .! "-'

r -1•:10- mr.1 .\f•;·M ) !2•?.t l~

1'(•.1'1111) 2ote)! 71

• U•11· 1) 011:::0.<n

f14inm·1) l ()-' 7.l?t0.91

•1'!( mi t1) ?t:.9::: 14.1

n,(ml mln) 1?4.6 .! ~I.I

4 (1) 11.?tS.5

\h·~c .. ·~m11 ~, 14 t 11.4?

r At (" I \ c " :!O•\.l • IO~k; 511t•- I

( \\G tSPl t \'· Ja ~al) ( \\CitSOJ

!H1t90 l'9).!11ti l.4 ?• 1 l :~U

4l.0.1. l l~ 10:.f11S4 , 16.l tnS

1'7.1'.! 10.7 46 4 1 l'-' ~:4 a 14.0

l "4S1,? '9?1110 llh4J\

lfi'74 .! 4.;(, '71fi 11 410_~ 41'7.<lt 19$2

11U: l l.7 !Y\lt21.SI II 5'16 4 1

OOll!O.Oll 0~:::0019 0~10010

: 9410_9} ? . St0<'9 t ~;;s l M

2~1-1 ! ~ ~ .\ 211<> 6 I 1~ l 2\:<t.~1 19

4~ llJ I 1 2.~;) 11S.h ' l .9 1~4 It 9' ..

14.,, 10..S ' -'": 10..s SS.1tl•'

1<1(1.:::'0 -.,_1_,1.\:::llL' 9 10"6t 4 .1'

0.0!>

00~·

O('••

••

o.o.· OM•

00$• .. , I!!J2.!s..!- Phannacokineuc parametm for CAPE and CA P.-\. Non­compartmcntal analysis (NCA) and bi-cxponrntial fi t 10 a 2 compartmcm model are shown

:_-,.,,~lusion.

- An HP LC-lvfS method for the quantitative dete1mination of CAPE and CAPA from rat plasma was developed and validated

- The validated method was used to dcte1mine levels of CAPE and CAPA following intravenous administration to m ale

Sprague-Dawley rats in a pha1macokinetic study

- CAPA exhibits a significantly longer elimination half life from the sys temic circulation than CAPE

- CAPA appears to exhibit non linear phan11acokinetics in the dose range of S - 20 mg/kg

Comrarison of caffcic acid phcnylcstcr (CAPE). caffcic acid phcnylamidc (CAPA) ~md 2-cyano-3. 12 dioxoolcana-1.9 dicn-28-imidazolide (CD DO-Im) in protecting human cndothdial cells

from O' · c Sl .. css: ·~"1l "i{olc of J a ox,gcnasc 1•2J. Bynum, 2P.D. Bowman, and 15. Stavchansky

1Division of Pharmaceutics, College of Pharmacy, The University of Texas, Austin TX 1US Army Institute of Surgical Research, San Antonio TX

Abstract

A munber of dmgs have been identified that produce a cytoprotective effect against oxidative stress that is correlated with induction of heme oxygenase (decycling 1;HMOX1 ). However, determining which is most efficacious in cytoprotection has not been addressed. CAPE, CAPA and CDDO-IM, cytoprotcctants that principally work via inductions of heme oxygenase (decycling l ; HMOXl) were compared.

Introduction

To better understand the protective properties of caffeic acid phenethyl ester (CAPE), caffeic acid phenethyl amide ( CAPA ). and 1[2-Cyano-3, l2-dioxooleana-1.9( 11 }-dien-28-oyl]imidazole (CDDO-IM}. a synthetic triterpenoid. an i11 1•ilro model of oxidative stress was investigated. Human umbilical vein endothelial cells (HUVEC) were treated with menadione (MD) as the stressor [I]. Cell viability and heme oxygenase-1 (HMOX l ) induction were monitored as a l>iomarkers of cytoprotection.

CAPE and CAPA are cytoprotective against menadione­induced oxidative stress in HUVEC COll"clating with the induction ofHMOXl gene, and its protein product HO· 1 [l ,2](Figures 4-6). The cytoprotective activity of CDDO­IM, a synthetic t1iterpenoid (a gift from Dr. Michael Sporn, Dartmouth University) against oxidative stress was investigated. Dose response studies indicated that CDDO­IM at 0.200 µM was more cytoprotective against menadionc toxicity than an optimal dose of CAPE (5 11M), resulting in endothelial cell survival of 80~o compared to 60% for CAPE.

Materials and Methods

l. Cell culture:

HUVEC (Lifel ine Cell Technology. Walkersville, MD) pooled from JO different donors were cultivated o n 96-wcll multi plates in MCDB 131 medium. Only the second thr ough fifth po pula1ion doubhnss of cells were used.

2. /11 vitro assay:

Celt viabilit) was assessed at 24 hours after initiation of treatmenl using resazurin (Sigma-Aldrich, MO). CAPE, CAPA, and CD DO· IM were assayed for cytoloxicily. Doses of CAPE, CAPA, and COOO-IM (0.5 µM and 5.0 µ M fo1 CAPE and CAPA and 0.200 µM for CDDO-JM) were delcrmined and used for 1hc cytoprotection assay. Confluent HUVEC were pretreated with various concentrations of CAPE. CAPA. CDDO-IM or 0. 1° 0 DMSO (control) for 6 hrs, then exposed to a tox.1c dose of MD for an additional 24 hrs. Cell viability was compared to the vehicle controls. At least three independent experiments were performed and each " as performed in quadruplicate .

3. H0-1 lnduction Confi rmation:

Dose response studies indicated that COOO-IM at 0.5µM was more cytoprotccllve agamst menad1onc toxicity than an optimal dose of CAPE and CAPA (5 µM\. resulting in endo1helial cell survival of 80% compared 10 18% for CAPE. Messenger RNA for HMOXI was increased 90-fold in the presence of CDDO-lm, while only 13-fold by CAPE (Figure 5) Westem blot enalyS1s of H0-1 indicated 1ha1 by 6 h. CDDO-IM induced an 87-fold hig herlevel ofH0-1 while CAPE induced a 13-fold increase (Figure 6) . The results ind icele that CDDO· IM is a more potent cytoprotec1ant than CAPEICAPA, and 1his beneficial effect correlated well with the induction of HO- I.

Results

---1 CAPE cytoto•icity in HUVEC (P31

110

! IOO '·11111 ! : ·- -- . . . . ··1 i '° .... ----·-3 lO • ·-- -

0 1-=1

to ~ ~ •o ~ n 100 l&MCA'l

Fig\lre I. Cytotoxicity of CAPE in HUVEC. • · p-r. 0 05 versus conlrol (0 ~1M CAP E). CAPE 3t doses or SO, 75, and l OO .,i M were C}1otoxic and less than lO .,iM \\ere used for C}10prole<:t1on ana~.

r--- -·---, CDOO-IMtclCltlty In HUVEC (Pl )

I "" t" 100

t ••. llllll II ~ "' f •o I I I! 20

I IO "' ""' ... 1'0 . ....

l_ _______ _::o~_:1 ______ _J

Figure 2. CytotoxiCll)I of CDOO-IM in HUVEC. • : p-::: O.OS \·crsus control (0 nM CDDO-IM). CDOO-JM :it doses of 750 :md 1000 nM were cytotoxic :ind lcu than 200 nM \\ ere used for cytoprolcction M U ) .

I- Crt~c.ti~OO·IM;1colinst7S idi"4t>~H~ IPJ) f

m -

J 100 . - ·1 8 "

I ~ 1 HI IL,, ~ tr. ~-

L l

,,-'j~//./l/.///i',l/./ e .l#'~.t'l// <TJ J , • .,.

Figure 3. Cyloprotcction of CAPE rutd CDDO·IM a.g:iinst 75 ~1M mcnodionc toxicitv in HUVEC. The cytoprotcct i,·c cfT.:et or CAPE nn.d CDDO·IM wM dose dcpcndcnl CAPE DI 20 ~Lf'..1 protected HU VEC ngat-nst MD·ind11ccd to~ich)' l-10'61. cell 5urvi\•.:1I) resu lting in nppl'oxinrnt;,:ly 111 n;. cell su1"lfrlll. CDDO·IM c.\hibitcd its' hi&hcst prolection against MD-induced lo'\ icily al 200nM rcsulling in oround ~0% cell sun j , :ii

Cytoprotiectlon o f HUV£C aa.a lnst70 uM M D

i I i I I Fign rc 4. Cytoprotccuon of CA PA, CAPA, an d CDDO· IM. Compounds gh·cn before MD induced mjul)' resulted in signifac:intl)' better cell SUT'\'h-:11 CDDO-IM provided t00-.4 ccll ,•iabili~.

!2 ,,. l

---• 6 I!!

l :!

~ .. "

100 •

0

ii " ~ . f .. l!

i "

•• r.n ..

10 .

~u • . ·-V?J::,Ot'""ll l--.~ \.LIV.)·IM

Fiiurc 5. HMOXI gene expression, mRNA inductlOtl in HUVEC b)' 6 hr 1rc21tmcn1 of~ µM CAPE and 200 nM CODO.IM. HMOXI RNA wH induced up to 90 (old h> CDDO·IM comp:sred lo :1 13-fold incrc~c followine CAPE treatment

100

90

80

70

GO

so

"' , . •• 10

DMSO <C ll) CAPE CODO.IM

Fis,urc 6. HO- I exprcn1on in HUVEC by 6 hr lfCollmcnt of S µM CAPE and 200 nM CDDO·IM. HO-I protein wos induced up to 87 fold b)· CDDQ.JM comp211-cd to a I 0-fold inerc:n:e following CAPE trc:Umenl.

c. fld s if>.!:!.~. I . (}1otoxicity pfofilcs of CAPE. CA.PA. ond CDDO-IM were established in

HVVEC. CAPE abo\·e 40 ~lM and CDDO-TM OY Ct SOO nM were C) IOloxic.

2. The cytoprotecth·c effects of CAPE. CAPA. and CDDO-lM were dose dcpcndcnL The cytoprotection ofCDDO-IM is much more potent than thal of CAPE and CAPA

3. The induction of HO-l by CAPE. CAPA. and CDDO-IM corrchucd well \\ ilh their C)10protcction.

~. CDDO-IM prod ded signilicnntly heller cytopro1cc1ion 1h:m bolh C' APE and CAPA as o. pos1 treatmenl to MD induced injlll)'.

S. Since CDDO-IM ;,ppc:n 10 pro,·idc im pro,·cd cytoprotcction ::igoinst ox1doti\·c st1-,;;u it is a good condidalc for testing in future models of ischcmia-rcpcrfosion inJu~.

Reference~

I \lo'~X.!1~ . ~-U-8. 8',-JA. i..-.... .Sl l a-.- PD <...,..,..l'lt~ • ._,,... e/ hwn;in~~«lklfMl~C)totoMnlyby...,ft .1. arid ~)t<'st-. 1hcnJk .,fhr""8 0 •\'J~I £ur Jl"t!.-_,.ol 2(JOt~~ 4.S'Oi(l ·l) ll-"H

l Y11ni;.. J f ,• / Srn1ho-.., lilf • .. ;,,.~' 1•f\:~ f!-.w 3'.'1<1 pl'wn<t1hyl .1m<•kl IC \ r ~)illl(>llMl•d

J.c.'i~1w• tOMf''' " ''"' of ~)til'fW"4""''"" rlralJ 10 c"'1i:ic :ktd ~lhyl •O(f f( 'APE) j-~MfiiCIHt 1 1.~~Z-~Jl,~ S~IOJOOSl9-S IP"JIO 101c.1bliM- l'OIOOS O' ·J f20 10l

Network Analysis of the Cytoprotcctivc Effect of CD00-11\1 against Oxidant Stress in Human •.J•nbilic~I Vein Endot"eJia' Cells (Hl'VEC)

1•2J. Bynum, 2P.D. Bowman, and 15. Stavchansky 1Division of Pharmaceutics, College of Pharmacy, The University of Texas, Austin TX 2US Army Institute of Surgical Research, San Antonio TX

Abstract

I [2-cyano-3, 12-dioxooleana-1 ,9( 11 )-dien-28-oyl)imidazole {CDDO-IM), a synthetic derivative of oleanolic acid. has been demonstrated to possess anti­inflammatory activity. CDDO-IM {gift from Dr. Michael Spom. Dartmouth University) was compared to the phenolic cytoprotectants caffeic acid phenethyl ester (CAPE) and caffeic acid phcnethyl amide (CAPA). CDDO-!M at 0.20 1tM was more effective than CAPE or CAPA at 500 and 5000 nM in protecting HUVEC from oxidant stress produced by menadione. Since CDDO-!M exhibits no direct antioxidant activity we tested it for transcriptional activation with whole genome microarrays and found that about 250 genes were up- or down­regulated by CDDO-IM. In addition to up-regulating. heme oxygenase- 1, a well-known cytoprotective gene, it also induced members of the heat shock protein famil y. Submission of genes statistically al tered in their expression by greater than two-fold up-regulation to Ingenuity Pathway Analysis (IPA) produced networks known to be related to cellular development. growth and proliferation. cell s ignaling. and canonical pathways including NRF2-mediated oxidative stress response and PPAR signaling indicating that cytoprotection involves multiple pathways in addition to the well described phase II enzyme induction.

'r:i.•.r.9_:.ci!,"QiO!,

Oxidative stress is commonly encountered in ncurodegencrative diseases such as Parkinson's, Alzheimer's, and Huntini,>ton's, vascular disorders including strokes and hea1t attacks as well as traumatic injuries. We previous!) reported that CAPE and CAPA displayed cytoprotective actidty against menadione (MD)-induced oxidative s tress in human umbilical vein endothelial cells (HUVECs). The induction of heme oxygenase-1 (HO- t ) , a phase II enzyme, in HUVEC played an important role for CAPE and CAPA cytoprotcction (1,2). To improve the beneficial effect. a more potent phase II enzyme inducer, I [2-cyano-3, t 2-dioxooleana- 1.9( 11 )-dien-28-oyl)imidazole (CDDO-!M). was examined. The cytoprotection mechanism was investigated through gene expression and signaling pathway analysis.

Materials and Methods

In vitro assay: Confluent HUVEC were pretreated with 0.5 µM or 5 µM CAPE or CAPA: 0.20 µM CDDO-IM; or control (0. t % DMSO) for 6 hrs, then exposed to MD for an additional 24 hrs. Celt viability was assessed using CettTiter Blue. Each experiment was performed in quadruplicate.

Gene expression ana lysis:

BRB Arrav Tools Total RNAs from 6h-CDDO-!M (0.20 µM) pretreated or vehicle control HUVECs were isolated and labeled for microarray analysis using Agilent whole-genome microarrays. The experiments were done in quadruplicate. Microarray data analysis and statistical comparison were perfom1ed using BRB Array Tools (http :Ill in us .nci. nih. govlBRB·ArrayTools. html) Genes were considered statistically signifi cant with P value< 0.00 I and FDR (false discovery rate) value< 10%.

Ingenuity Pathway Analysis Genes significantly altered in their expression were submitted to Ingenuity Pathway Analysis {IPA) for further network analysis (www.ingenuity.com). IPA used Fischer's exact test to calculate a p-value determining the probability that each biological function and/or disease assigned to that data set is due to chance alone. The score of network represents the negative log of the P value. Therefore. scores of 2 have at least 99% confidence of not being generated by chance atone. Genes are represented as a single node in the network. The intensity of the node color indicates the degree of up- (red) or down- (green) regulation.

Results Cytoprotect lon of HUV[C •piMt 70 uM MD

I .. l

i I i I I Fiaurc I. C)1oprolccoon by CAPA, C'APA. and CDDO­IM aiamst menachonc induced oxidatnc sh-est. A six hr pretrc:umcnt before MD iaduccd mjury rcsnltcd in si,gnifican1ly beltcr cell S\lf\"i\'a l. CDDO-IM provided lOO"/.ecU\·1abil1ty

Table l : Most mopped up- and dov.11-rel,lulalcd ~encs by CDDO-IM through IPA:

""~, .. ~~ Joliil~ I! · ....

'°'"',.,,.....~,.··" ~r.,.,..tr ~riolJ11 1llffQJI ~JltW'Mll {NM.flJ:IW'I! m t'lt-<0§ o.cou ht1I 11\oct; 10li:01 P!'Olt •~1A IHS'•W INM.005J'5J "'' t9«-n 0.0057 •Uklat1¥t' ~t,. 1 11Mucdcro-th flilt.ibllor I •OSGINU, lf~,1 ¥fflllf'I~ L [MM.OUJ10[ ,,, •Uf·Qlj 0.0C?I

"tMOJVJC-llcc,dl"C)l(l-W()Xtl(~~IJ· u U•~ o.mr tn.NIVMM\ i•tl•llJMll:t " OOCl)MJ o.om ~e ... -.-....-~o.,,...,,1.111t111MtlfMftt>.lliM,.P:01m ,, t 10fQlj o.an ~~lf"IJl.qtlll~wMtrNtrA.mctnbu 4UlffAJA4J 1HM_Ollf.Oll " o~u o~m

rWrutMVltttM Ila~~. ,,.odlf1t11ubUftltjCiC.l.MI, INM_QjX(jlJ 6.0 0 0Xlll7 O.CO'JI

~- ......... ~J(J.llC)J. jtt M.l•Mll .. O_ccetl1 O.OISO

Wan.,..rttMl_ ISm.llO!J. trMKlll!!'fflllllll 1.jtiM l:ll'!!i .. D.000114

rc,lOJM)fff~pUM~

•·1mi1Mull ~Wo!l lMI«'* I ;VU."4!1, tr•llWlfl .,.~JN L [NM,JOI0711 ·l-U u oc.os 0.0056 b;l(ut~ot IAP ftjlUf•rllllll!ll'"" J(WICll 1r111Wlpt vlll•~I t, [HM,..a:nm: ·11.7 ()NJO?S 0[)111

11~111 f(tl'ldolhll!li.l .tdlltHll9'GlfWW!J) ISl"W IHl.(..«0450I ·1l1 O.CO>JU D.01'10 """tlt, ....... t1111llWU. 1tM&U11J11....,.,..l INMJll'.l~I ... tOCOlll 0.0140 ~~M!Wl_.,.._.,rl(OHISJlilfM,.COl?S. . .. l lJC-<IS o.COK :wtorlvll!MN~flll''ltlyJt.HM.-.,1,po;,.,,rtldt! (Cl'tHlll IN"l.OI~] u O.DODU C.t:lm N•ll(flt'tUl".-ltr11• - iNATl),[HM,J.IO)""..«!I ·S. I O.C'O.llJl C.C091 IM-do1ll lflliH~Jll011:1'ilTXlf\1l,[HMJl0641:1 .. o.oom1 o..ouc 1nwb~·..-•ivowthfackW~1tett!llilCJW11, 11-'0!¥M.ltlll.l "li.t_~ ..., f~Q.OlM

ltudMP'dlNfktl~"C.PftWll).rNM_~~UI •. , ot.w:»r.a emu

'OO lc l : Top rcgufotcd biological networks b) CDDO-IM through IPA (scores ~ 14): II ._.. __ _

11:• n1.n• .. •~.•U11-M1t • 1, ~•--~~•t..._...., • .,.;111. l •1 .. ......... ~-1·-·-_,.. ...... 1_.,,......, ... llN.-4...iUOU/l "'°"I #ftl.'• Ow:D l "ltU.-..U. .... • -.S.

•..-.·-~-.n-t OAll,.• ..-,<---.,.... ...... _, tC9W.~• •- .. Ii. ·~"'"~.1 ............. _..r>Q-1 ~ o. ... , .... ~,.... '"""~~ ...... '"""" '~·-·"""'--·· " ' '"""' ,,,_., ,.,,,, ,_ .. " .....

J , ..,,.., _ _. .... .. ' fW"'"-<Jlf'llo Ol. :0-""IHU

.., --..--·· IOCJ. ... 1 . . ...... IGXlt?\_,.

':.,\ " :;"".,.:..~-::~~==-"-...... ~ 1-w.. •o.--.. 111Wt••WJG.-"~ ' •-u. r .w..U.-UOI\ te-,tt · - l.•:..Mf', - 1'-1'K f Jlllf, ..,.ltW")W'I· _.., _ ,

;%.__,..•:.·~·.~:. :;;: •-.u-.11, •-.•~m;;'.Jll:• "'-I , ...., . .... ...:- •:. ·mlQ.~-'([IC~~ICll:, ICQ.,t'l.. IU

........ .. ~ -.i ·~ .. ~- ...... , . .... . ., """' , ....... ...

';;"" ' .......... ·-· ~" ........ . . -. ........

I

I ._,......,.~M1")ill.G'""<Vl:._.., __ ,.._J. IMJC. ......... l:i • 111. ,..m.w.-.ur.r ,,c.11~..,,~-"., .,."".> .. Hn _.._ "<OP"\ \ 'f .., , l "-'Ql.'lfU-- d • l,q .. t• 1.•ut .•1& +1tRa1 A !IC~ • • , ... , ... ,. _ '""~ ·-

~·""""'...-..~<~r­n r_.ra

c.,.--., __ ,....,_..,. ·-·--~-....-..--. ..... - ...-.:.i...- (oh., ....... .... = ....... ,...-~ ··~'"""-.; ........ """" ......... -

..,.,,,.· - ·-·····~~~···~ 1· ---- ··--. .-...... ~ui..-- .. --..·-·- - t...-;«1-•c:.·.•..._,",_,. ,._.,.,.. _......., ~_,, ... .-. ...... ~ .... ........ -. ... -_._ ~ •"·" .. ._U -1.llH"~•O.-"'- ' "'" H _.._,:.~c.i..- t.oc'":tOl __

1.

~ ..... - ..... .,_,.,,.. __ e~-..· -.-.. r1,.1 """·•it·l11' ·' ~-•~- - p - 11.•r·wr ... ., .. ---· ·----. ----- ~- -- -·---- ..

I.

2.

=:d;_-:;:..::...-= ..:..=::"=..:..

~

-,, 1. :-.-~ ;, -·· ..... _.... -''··

• h I • • ~· .... r .,...., ,...,. .. , _

-:::::-=-- - -·- ~,.

·- :·

_, - - .....

! =-

_ ... ;.

_-;..'.

;~:~<? "';i' ~::.~-~~ ~:-; ._

',.\,, ,_ ,..,. ,_

-- ,/ , •, ~

•• ·~ i, - .,, -;;.-r-:: '·I~ ~ I ~·

\ · \·~ ',, -~

~/ ~:-1 :;:/ -:·•

:. .:...

..; .!,

"':':'~=·

-=-:-=· - -

FtilH l l'\W.?-.liawl "\ti.11111• .. llro• 'isNl111t r'11t,._-~- 1 1i.M;d to.· ("DOf>-IM I~ (l'\">l11t• l ,\ '< I O-'\. "JlMNd ,olor loltlll~rtpl'«<lllSl!f'-Nl\lla1cJr" ·•

~~Ju s'_ ns CODO·IM is ;i. more potent cytoprotcet:i.nt lh:m CAPE or CAPA 3g:i.inst MD· induccd o.xidoti,·c stress in HUVEC in lhc mcn:Kl ionc model Gene expression profilin' and signaling pathv.~· analysis indicaccd tha1 NRF2 mcdi;itcd p:1lh\\I) 1s ;,cti \':lted by CDDO·IM, which confirmed previous find ing in lhc li1cr:itur-c (3). Other sianolins pnthwoys :mociatcd \\i lh CDDO-IM trea1mcn1 include PPAR and xcnobiot ic mct3bolism. These ruuhs indie:1tc 3dd1tion31 l:lf"&clS lh:1t may uplain 1t.s bcUcr ~1oprotcction ..

Bderc !)..<:~-I l\~X.S... . ...... •$ ZllaooD e-J..\.~Pl8""-pn "-"'P*-'-"' ......

.....,."".__.._,, ........ ~i., ....... -.t......,..,1- ...... .,.._""''CC--·· I E>wJ...,_1ol ~Sqt ~ ffl(l•1';,1.)S

1 , . ..,,,, ~' "' s,,~i... ... r>1 • .....,.,,a1 .._. ;r..._ .., ,.,.~1 ,n11NrC'.\f'') t1....,,,._;111.,. *''·"'"""" ~.,.llf,•1"'111f-h,. 1ll..c1uo"fTcic w• ,i,c11e1h"h-U1 (l'\f'l'- • - 1 •,1r~ ... 1•.Sfl'l­~cn.t. • t :~~""<l•IOOSlt-J IPillDIOl6).lolllClO\OOS.lllOllOIA) .

.l u.., . ..:tt -n.~.-w..~("tl()C)_.~.,_,_......_..,._

-.~1 ...,~~-\Rl.: .... "f.C"an · tn'1. l~~-'i 11 .r.tJfptltl•11U:'CIDBS­~~1lCA""'°" ..... \'t,~)

Comparison of Bioluminescence Imaging and Luminometry for Detection of Luciferase

Activity in Transgenic Mice Engineered to Express Luc if erase in Response to Hypoxia

Ashish Rastogi 1.2, James Bynum 1•2

, Salomon Stavchansky2, and Phillip Bowman 1

1 US Army /11s1i1111e of S11rgica/ Rcscarclt , For/ Sam Ho11s1011. TX 78234 1 Division of Pltor111ace111ics. College of Pltormocy, The L'11il'ersity o,{Texos at A11s1i11. TX 78712

Abstract: Bioluminescence imaging was compared with luminometry for quantitative determination ofluciferase activity using transgenic mice as a model that is engineered to accumulate luciferase in response to hypoxia. Mice were hemorrhaged and in vivo imaging was performed at 4 h. Mice were then euthanized and organ reimaged ex vivo or frozen prior to luminometiy. Luminometry rather than imaging showed differential expression of luciferase in hemorrhaged mice compared to sham indicating upregulation oflllF-l a. Luminomet1y was found to be more precise than in vivo or ex vivo imaging for detennining the effect of hypoxia for this particular mouse strain.

©2010 Optical Society of America

OCJS codes: ( 170.0 110) Imaging Systems; ( 170.3880) Medica l and biologica l imaging.

References and links

I. A. Sato, B. Klaunbcrg, and R. Tolwani, " In vivo bioluminescence imaging," Comp Med 54, 631-634 (2004).

2. Y. Kong, Y. Shi, M. Chang, A. R. Akin, K. P. Francis, N. Zhang, T. L. Troy, H. Yao, J. Rao, S. L. C irillo, and J. D. Cirillo, "Whole-body imaging of infection using bio lum inescence," Curr Protoc Microbiol Chapter 2, Unit2C 4 (20 I I).

3 . R. T . Sadikot and T. S. Blackwell, "Bioluminescence: imaging modality for in vitro and in \i vo gene expression," Methods Mol Biol 477, 383-394 (2008).

4 . C. H. Contag and M. H. Bachmann, "Advances in in vivo bioluminescence imaging of gene expression," Annu Rev Biomed Eng 4, 235-260 (2002).

5. P. E. Andreolti, G. Caceres, R. Zamkina, and M. Dauphinee, "Luminescence applications for chemotherapeutic drug development," Recent Resu lts Cancer Res 161 , 3-12 (2003).

6 . A. Moller and J . K. Jansson, "Detection of firefly Juciferase-tagged bacteria in environmental samples," Methods Mol Biol I 02, 269-283 ( I 998).

7. G . R. Rct1 ig, M. McAnuff, D. Liu, J. S. Kim, and K. G. Rice, "Quantitative bi oluminescence imaging of transgene expression in vivo," Anal Biochem 355, 90-94 (2006).

8 . G. R. Reuig and K. G. Rice, "Quantitative in vivo imaging of non-viral-mediated gene expression and RNAi-mediated knockdown," Methods Mol Biol 574, 155- 171 (2009).

9. N. Porakishvili, J. L. Fordham, M. Chanel , P. J . Delves, T. Lund, and I. M. Raitt, "A low budget luminometer for sensitive chemiluminescent immunoassays," J lmmunol Me1hods 234, 35-42 (2000).

10. P. E. Andreotti and F. Berthold, "Appl ication of a new high sensi1i' ity luminometer for industrial microbiology and molecular biology," Luminescence 14, 19-22 ( 1999).

I I . M. Sa fran, W. Y. Kim, F. O'Connell, L. Flippin, V. Gunzlcr, J . W. Horner, R. A. Depinho, and W. G. Kaelin, Jr., "!\louse model for noninvasive imaging ofH IF pro lyl hydroxylase activity: assessment of an o ral agent that stimulates erythropoietin production," Proc Natl A cad Sci U SA 103, 105-110 (2006).

12. H. M. Sowter, R.R. Ra, al, J. W. Moore, P. J. Ratcliffe, and A. L. Han is, "Predominant role of hypoxia­inducible transcri ption factor (H if) -1 alpha versus Hi f-2a lpha in regulation of 1he transcriptional response 10 hypoxia," Cancer Res 63, 6 130-6 134 (2003).

13. K. K. To and L. E. Huang, "Suppression of hypoxia-inducible fac1or I alpha (HIF-1 alpha)1ranscriptional activ ity by 1he HIF prolyl hydroxy lase EGLN I ," J Biol Chem 280, 38102-38 I 07 (2005).

14. H. L. Foster, .I. D. Small, and J. G. Fox, The Mouse in Bio111edical Research: Normative biology. im1111111ology, and h11slx111d1y. Americau College oflabora101y (Academic Press: New York., I 983), Vol. Ill.

I 5. S. Gross, U. Abra ham, J. L. Prior, E. D. Herzog, and D. Piw nica -Worms, "Continuous delivery of D­luciferin by implanted micro-osmo1ic pumps enables 1rue real-time bioluminescence imaging ofl uci ferasc acti\'ily in \'i\'o," Mo l Imaging 6 , I 2 1-1 30 (2007).

16. J. J. Lysiak, J. L. Kirby, J . J. Tremblay, R . I. Woodson, M. A. Reardon, L. A. Palmer, and T. T . Turner, "Hypox ia-inducible factor-lalpha is constituti vely expressed in murine Leydig cells and regulates 3beta­hydroxysteroid dehydrogenase type 1 prom oter activity," J Androl 30, 146-156 (2009).

17. S. C arlin, A. Pugache\", X. Sun, S. Burke, F. Claus, J. O'Donoghue, C. C. Ling, and J. L. Humm, "In \ivo charac1eriza1ion ofa reporter gene sys1em for imaging hypoxia-induced gene expression," Nucl Med Biol 36, 82 I -831 (2009).

18. D. A. Chan, P. D. Sutphin, N. C. Denko, and A. J. Giaccia, "Role ofprolyl hydroxyla1ion in oncogenically stabilized hypoxia-inducible fa c1or-I alpha," J Biol Chem 277, 40 112-40 117 (2002).

19. G. F. Lungu, M. L. Li, X. Xie, L. Y. Wang, and G. Sto ica, " Jn vi' o imaging and characterization of hypoxia-induced neovascularization and tumor invasion," Int J Oncol 30, 45-54 (2007).

20. S. Salceda and J. Caro, "Hypoxia-inducible factor I alpha (HlF-1 alpha) prote in is rapid ly degraded by the ubiquitin-proteasome system under nonnoxic conditions. Its stabilization by hypoxia depends on redox­induced changes," J Biol Chem 272, 22642-22647 (I 997).

21. R. J. Viola, J . M. Provenzale, F. Li, C. Y. Li, H. Yuan, J. Tashjian, and M. W. Dewhirst, "In vivo bioluminescence imaging monitor ing of hypoxia-inducible factor I alpha, a promoter that protects cells, in response to chemotherapy," AJ R Am J Roentgenol 191, I 779-1 784 (2008).

1. Introduction

Bioluminescence imaging is a simple and sensitive method that is based on detection oflight emission from cel ls or tissues [l ]. The luciferase gene is commonly used as a reporter under the control of a promoter of interest. The technique provides a low-cost, non-invasive, and real-time method to perform gene expression assays in living animals [2-3]. The technique greatly reduces the number of animals sac1ificed per experiment while providing sufficient information on gene regulation and protein function in the context of functional ti ssues and organ systems [4]. It has been widely used to track tumor cells, bacterial and viral infections, and gene expression and to develop chemotherapeutic drugs [5-6]. Additionally, quantitative in vivo bioluminescence imaging has been used to study the efficiency of gene transfer in the liver and other organs [7-8]. Luminescence from the expression of the luciferase gene can also be measured in vitro by using a luminometer, which provides a convenient, rapid, and sensitive method for quantifying gene expression when used with a luciferase reporter assay system [9-1 OJ. However, the two techniques - bioluminescence imaging and luminometry - have not yet been compared to determine which provides a better quantification of luciferase activity. In the present study, bioluminescence imaging and luminomet1y techniJ,ues were compared using a transgenic FVB.129S6-Gt(ROSA)26Sor11111fHIFJ tluc;Kael/J mouse strain that has been genetically engineered to express luciferase with accumulation of HIF-la [11]. Excessive loss of blood, such as in hemotThage, causes reduced oxygen supply to some organs leading to accumulation of hypoxia inducible factor (HIF-la) [12-13]. However, those organs most effected have not yet been determined. In this study we have attempted to use in vivo imaging to determine the luciferase activity as a function of hypoxia in organs most affected by hemorrhage. Organs were removed and imaged again ex vivo. Finally, organs were homogenized, and luminescence was measured with a luminometer. The results of the three studies were compared to determine which method provided the most precise way to quantify luminescence from hemorrhaged organs.

2. Materials and Methods 2.1 Mice and Hemorrhage

Experiments were conducted in accordance with the guidelines set forth by both the US Army Institute of Surgical Research (USAISR) and the guidelines of the National Institutes of Health (NIH) for animal care and use. The study was approved by the USAISR' s Institutional Animal Care and Use Committee. Male FVB.I29S6-Gt(ROSA)26-Sor/1"1rHmA luc)Kaell.1 mice (Jackson Labs, Bar Harbor, ME) 8-12 weeks old and weighing 25-30 g were used in the study. The mice were allowed food and water ad libitum and provided with environmental ernichment tools. Before the study, they were observed for I week to allow for environmental changes and to exclude the possibility of pre-existing disease.

Animals were divided into two experimental groups, sham and hemorrhage. After anesthesia with 2% isoflurane in air, hemorrhage was accomplished by removing 40% of the calculated blood volume wi th a lancet (Medipoint, Mineola, NY) via the submaxillary vein of the anesthetized mouse, and sterile gauze was applied to the vein to stop further bleeding. The calculated blood volume to be removed for each mouse was based on its weight [14]. For sham controls, lancet was applied but bleeding was immediately stopped by application of sterile gauze.

2.2 Imaging

A 50-mg/ml solution of potassium salt ofD-Luciferin (Caliper Life Sciences, Hopkinton, MA) was prepared with phosphate buffered saline, pH 7.4. Continuous delivery of luciferin was achieved by using osmotic pumps (Alzet, Cupertino, CA) as described by Gross et al. [15). At least 24 h p1ior to imaging, the osmotic pumps were filled with luciferin solution and implanted on the dorsal side of the mice. Four hours after hemorrhage, mice from both groups were anesthetized with 2% isoflurane-air mixture. The mice were then imaged with the In Vivo Imaging System (IVIS<rl) Lurnina II biolurninescence system (Caliper Life Sciences, Hopkinton, MA). The light coming from various organs, as a function ofluciferase production concomitant wi th HIF-la induction, was quantified from the images with Living Image software 3.0.4. Mice were then euthanized and portions of each organ reimaged or frozen in liquid nitrogen and stored at -80° C for in vitro luciferase quantification with a luminometer.

2.3 Lumi110111eter Analysis

The luciferase assay system (Promega, Madison, WI) was used for quantitative analysis of tissues from the sham and hemorrhage groups. The analytical method was developed in accordance with the manufacturer's protocol. Briefly, tissues from both groups were homogenized in 500 µI of lysis buffer that was supplied as part of the luciferase assay system and previously mixed with Ix proteinase inhibitor (Thermo Fisher Scientific, Waltham, MA). The tissue lysates was centrifuged at 14000 rpm for 5 minutes. The supernatant was collected, and an aliquot was assayed for luciferase activity by using the modulus microplate luminometer(Promega, Madison, WI) and luciferin as the substrate. The light intensities were calculated by measuring the relative luminescence unit (RLU) signal. A portion of the supernatant was also used for determining the amount of protein in the ti ssue lysates with the Pierce 600-nm Protein Assay Kit (Thermo Fisher Scientific, Waltham, MA). The luminescence emitted from each organ was recorded as RLUs per milligram of protein.

2.4 Statistical A1talysis

Levene' s test was used to access the homogeneity of variance. Student's t-test was used to analyze differences between the sham and hemonhage groups. A difference ofp value < 0.05 was considered significant.

3. Results

3.1 Qua1ttitative Analysis of Organs Affected by Hypoxia by in vivo Bioluminesceuce Imaging

Jn vivo bioluminescence imaging has the potential to detect and quantify the expression ofHIF-la. The results of imaging FVB.129S6-Gt(ROSA)26 -So/1111<HIFfo?uc) KaelJ

mice are shown in Fig. I. The images indicate that overall the bioluminescence in the hemonhage group is higher than in the sham group. However, in most cases, it was difficult to determine luminescence from a specific organ and to study luminescence from the images themselves. For example, kidneys on dorsal images and testes and liver on ventral images were more easily quantified than the other organs. In Table I, average bioluminescence intensity (in counts) of these organs was estimated from the in vivo images of the hemoJThage and sham groups. Theoretically, organs from the hemorrhage group should depict more luminescence as a function ofHIF-la activity due to hypoxia. The testes from the hemonhage group exhibited lesser bioluminescence than from the shan1 group, the liver showed higher bioluminescence, and the kidney showed no significant difference between the groups (p > 0.05).

Pre-hemolThage 4h Pre-hemorrhage 4h

~ • m Dorsal

Ventral II Sham Hemorrhage

Fig. I. Non-i nvasive images of FVB. I 29S6-Gr(ROSA)26 -Sor""'""FI" '"J!'"' 1' mice (a representative mouse from each

group} are shown. A reference intensity scale with units in counts or photons emitted is also included.

Table I: Average bioluminesccncc intensity (in counts) of organs as observed from the in vivo images of the hemorrhage and sham groups are shown. Quantitative analysis of kidney, liver, and testes is shown as accurate positioning of other organs was not visible. The data is represented as mean with standard deviation (N = 3; *p

<0.05).

Group Testes Liver Kidney

Hemorrhage 165.4 ± 60.2 117.8 ± 62.9 54.8 ± 45.5

Sham 359.6 ± 53.6 74.6 ± 32.4 47.0±23.5

Ratio (H/S) 0.5* 1.6* 1.2

3.2 Ex vivo imaging of Organs

Bioluminescence imaging of the whole animal may be affected by various factors such as blood flow and scattering of light photons. Hence, ex vivo imaging of isolated organs was performed to observe the luminescence in individual organs. Fig. 2 shows ex vivo images of various organs isolated from both the sham and hemo1Thage groups. It is our observation that the intensity ofbioluminescence exhibited from an organ from animals belonging to the same group was variable. In addition, not all organs exhibited bioluminescence in each animal.

Sham Hemorrhage

Fig. 2. Images of various organs after removal from a sha m or hemorrhage animal. Images are from a representative mouse from each group.

3.3 Quantitative Lumillometer Analysis of Homogenized Organs

After homogenization, isolated organs were quantified wi th the luminometer. Table 2 shows the average luminescence/mg of protein of organs isolated from the hemorrhage and sham groups. The luminometer analysis shows that the hemorrhage group has higher luminescence values than the sham group in all sets of organs (p < 0.05). The results obtained with the luminometer are more in accordance with the theory that animals with hemorrhage due to hypoxia will exhibit higher luminescence than the sham group because ofupregulation ofHIF-la

Western blot analysis for HIF-l a in proteins from nuclear extracts of various organs such as lung, her, kidney, and spleen failed to yield any significant results. Our results were in accordance with a previous study by Lysiak et al., who were only able to detect HIF- la in testes [16]. One reason could be the low density ofHIF-la in the organs. In such a scenario, the in vitro luminometer provides a reliable and sensitive tool to detect fllF-l a in hypoxic organs.

Table 2: Average luminescence (in millions relative luminescence units/ mg of protein) of organs isolated from

hemorrhage and sham groups are shown. The luminometer analysis shows hemorrhage g roups have higher

luminescence values than the sham group indicating hypox.ia in the hemorrhage group. T he results are shown as

mean with standard deviation (N = 4; • p < 0.05).

Group Lung Liver Kidney Spleen

Hemorrhage 3.2 ± 0.9 13.7 ± 3.2 5.4± I.I 4.1 ± 1.2

Sham I.I ± 0.4 7.6 ± 2.3 1.7 ± 0.3 3.4 ± 1.7

Ratio (HIS)• 3.0* 1.8"' 3.2* 1.2

•Hts = hemorrhage.'sham group.

4 Discussion

Although in vivo bioluminescence imaging and in vitro luminometer analysis have both been previously used to study hypoxia-related luminescence [17-21 ], to our knowledge, there hasn ' t been a single report that discusses the correlation between the two techniques. In the present study, we employed a transgenic mice model to compare the two techniques for their ability to quantitatively determine the degree of luciferase activity in response to hypoxia. Hypoxia was induced by subjecting the Rosa-Luc mice to 40% hemorrhage. The resulting induction offllF-la in va1ious organs was studied in co1Telation to the amount of luminescence emitted. In vivo bioluminescence images demonstrated that fllF-1 a induction was greater in the hemorrhage group than in the sham group. However, bioluminescence imaging showed poor quantitative ability, possibly because quantitative differentiation of

different organs using in vivo bioluminescence imaging is difficult in conditions such as hemorrhagic shock, which affects various regions in a hemorrhaged animal. Different organs may also exhibit luminescence differently due to their positioning and distance from the light source. On the other hand, the luminometer provided a simple and rel iable tool to quantify the amount of luminescence emitted from several organs. Nevertheless, one clear advantage of bioluminescence imaging over luminometer analysis is that the former is a total noninvasive technique, whereas the latter requires euthanization of the animal. Bioluminescence imaging is an attractive method for qualitative analysis as it enables serial and rapid collection of data .. However, for quantitative dete1minati on ofluciferase activity, luminometry provides a more precise method.

5 Conclusion

Luminometer data rather than in vivo or ex vivo imaging analysis confi1111ed that hemorrhaged animals show higher luminescence due to accumulation ofHIF-la in specific organs. Some affected (hypoxic) organs did not yield significant amounts of light by in vivo imaging. The luminometer was found to be more reliable than bioluminescence imaging methods in the eYafuati on of hypox ic organs. The Rosa­Luc transgenic mice provided a useful model to determine luciferase activity in organs affected by hypoxia.