SYNTHESIS AND CHARACTERIZATION OF LANTHANIDE CHELATES FOR BIOMEDICAL IMAGING by JOHN MICHAEL MANSFIELD GRIFFIN, B.S.

A THESIS IN CHEMISTRY Submitted to the Graduate Faculty of Texas Tech University in Partial Fulfillment of the Requirements for the Degree of MASTER OF SCIENCE

Approved

Accepted

August, 2000

SO^

ACKNOWLEDGEMENTS

/^O. /O ^ -Cy'

First and foremost, I would like to extend my gratitude and thanks to my mentor, Dr. Darryl J. Bomhop, for his endless patience and support. Most importantly for providing an atmosphere that facilitated scientific exploration which allowed me to grow and learn as a scientist. I would also like to thank Dr. John Marx for his insight and time for which allowed me to excel in solving the many problems that I encountered. I would like to extent my deepest gratitude to Steven Roberts, Steven Hagedom, Kelly Swiney, and Mike Houlne for their friendship and support during this entire experience. I would especially like to thank Tim Goebel for his assistance in the synthetic and analytical endeavors described in this work. I would also like to thank my parents for their love and support. Last but not least, I would like to thank my wife, Arleen, for her support, love, and faith that I could do it no matter how discouraged I became during my graduate career.

11

TABLE OF CONTENTS

ACKNOWLEDGMENTS LIST OF FIGURES CHAPTER I. II. INTRODUCTION SYNTHESIS AND CHARACTERIZATION OF Tb-[N-(2PYRIDYLMETHYL)-N',N",N"'-TRIS(METHYLENEPHOSPHONIC ACID BUTYL ESTER)-1,4,7,10-TETRAAZACYCLODODECANE 2.1 Discussion 2.2 Synthesis 2.2.1 Synthesis of N-(2-pyridylmethyl)-l,4,7,10tetraazacyclododecane 7 2.2.2 Synthesis of N-(2-pyridylmethyl)- N', N", N'"tris(methylenephosphonic acid butyl ester)-1,4,7,10tetraazacyclododecane 10 Complexation with terbium or europium

ii viii

8 8 11 11

12 15 16 16 19 24 24 25 28

2.2.3

2.3 Spectroscopic Characterization 2.3.1 2.3.2 Discussion Results

2.4 Specificity as a Diagnostic Agent 2.4.1 2.4.2 Discussion Results

2.5 Experimental Procedures

111

2.5.1

General procedure for the synthesis of N-(2-pyridylmethyl)-N', N", N"'-l,4,7,10tetraazacyclododecane 7 General procedure for the synthesis of N-(2-pyridylmethyl)- N', N", N'"tris(methylenephosphonic acid butyl ester)1,4,7,10-tetraazacyclododecane 10 General procedure for the chelation of N-(2-pyridylmethyl)- N', N", N'"tris(methylenephosphonic acid butyl ester)1,4,7,10-tetraazacyclododecane with terbium 3 General procedure for the determination of absorption and emission wavelengths General procedure for the determination of the fluorescent lifetimes General procedure for the determination of biospecificity

28

2.5.2

28

2.5.3

29

2.5.4

30

2.5.5

30

2.5.6

32 36

2.6 Conclusions III. SYNTHESIS AND CHARACTERIZATION OF 6-SUBSTITUTED 2QUINOLYLMETHYLENEPOLYAZAMACROCYCLOPHOSPHONIC ACIDS AND COMPLEXES WITH TERBIUM AND EUROPIUM 3.1 Discussion 3.2 Synthesis 3.2.1 3.2.2 Synthesis of 6-Substituted Quinaldines 18-20 Synthesis of 6-Substituted 2(Chloromethyl) Quinoline 25-27 Synthesis of N-(6[substituted]-2-quinolylmethyl)1,4,7,10 tetraazacyclododecanes 28-30 Synthesis of N-(6[substituted]-2-quinolylmethyl)N', N", N'"-tris(methylene phosphonic acid)-1,4,7,10tetraazacyclododecanes 34-36
IV

38 38 40 40

41 42

3.2.3

3.2.4

43

3.2.5

Complexation with europium and terbium 37-42

45

3.3 Characterization of Lanthanide- N-(6[substituted]-2quinolylmethyl)- N', N", N"'-tris(methylene phosphonic acid)1,4,7,10 tetraazacyclododecanes 37-42 3.3.1 3.3.2 Discussion Results 46 46 47 54 54

3.4 Experimental Procedures 3.4.1 General procedure for the synthesis of 6-Fluoroquinaldine 18 3.4.2 General procedure for the synthesis of 6-fluoroquinaldine N-oxide 22 General procedure for the synthesis of 2-(Chloromethyl)-6-fluoroquinoline 25 General procedure for the synthesis of N-(6-fluoro-2quinolylmethyl)-1,4,7,10- tetraazacyclododecane 28 General procedure for the synthesis of N-(6-fluoro-2quinolylmethyl)-N', N", N"'-tris(methylene phosphonic acid)-1,4,7,10 tetraazacyclododecane 34 General procedure for the synthesis of 6-chloroquinaldine 19 General procedure for the synthesis of 6-chloroquinaldine N-oxide 23 General procedure for the synthesis of 2-(Chloromethyl)-6-chloroquinoline 26 General procedure for the synthesis of N-(6-chloro-2quinolylmethyl)-1,4,7,10- tetraazacyclododecane 29

55

3.4.3

55

3.4.4 3.4.5

55

56

3.4.6

57

3.4.7

57

3.4.8

58

3.4.9

58

3.4.10 General procedure for the synthesis of N-(6-chloro-2quinolylmethyl)-N', N", N"'-tris(methylene phosphonic acid)-1,4,7,10 tetraazacyclododecane 35

59

3.4.11 General procedure for the synthesis of 6-methoxyquinaldine 20 3.4.12 General procedure for the synthesis of 6-methoxyquinaldine N-oxide 24 3.4.13 General procedure for the synthesis of 2-(Chloromethyl)-6-methoxyquinoline 27 3.4.14 General procedure for the synthesis of N-(6-methoxy-2quinolylmethyl)-1,4,7,10 tetraazacyclododecane 30 3.4.15 General procedure for the synthesis of N-(6-methoxy-2quinolylmethyl)-N', N", N"'-tris(methylene phosphonic acid)-1,4,7,10 tetraazacyclododecane 36

59

60

61

61

61

3.4.16 General procedure for the chelation of the N-(6[substituted]2-quinolylmethyl)-N', N", N"'-tris(methylene phosphonic acid)-1,4,7,10 tetraazacyclododecanes with terbium or europium 37-42 62 3.4.17 General procedure for the determination of quantum efficiency 3.5 Conclusions IV SYNTHESIS OF N-(6-FLUORO-2-QUINOLYLMETHYL)- N', N", N'"TRIS(METHYLENE PHOSPHONIC ACID BUTYL ESTER)-1,4,7,10TETRAZACYCLODODECANE AND COMPLEXES WITH TERBIUM AND EUROPIUM 4.1 Introduction 4.2 Synthesis 4.2.1 Synthesis of N-(6-fluoro-2-quinolylmethyl)-N', N", N'"tris(methylene phosphonic acid butyl ester)-1,4,7,10 tetraazacyclododecane 43 Complexation with terbium and europium

63 64

66 66 67

67 68 68 72

4.2.2

4.3 Spectroscopic Characterization 4.4 Experimental Procedures

VI

4.4.1

General procedure for the synthesis of N-(6-fluoro-2quinoIylmethyl)-N', N", N"'-tris(methylene phosphonic acid butyl ester)-1,4,7,10 tetraazacyclododecane 43 General procedure for the chelation of N-(6-fluoro-2quinolylmethyl)-N', N", N"'-tris(methylene phosphonic acid butyl ester)-1,4,7,10 tetraazacyclododecane with terbium or europium 44-45 General procedure for the determination of absorption and emission wavelengths General procedure for the determination of molar extinction coefficient

72

4.4.2

73

4.4.3

73

4.4.4

74 74 76 78

4.5 Conclusions V CONCLUSIONS AND FUTURE PROJECTS

LITERATURE CITED

vu

LIST OF FIGURES

1. 2. 3. 4. 5. 6. 7. 8.

Pyclen-Based Terbium Complex Absorbance Scan of Tb-PCTMB Emission Spectrum of Tb-PCTMB Target Molecules Cyclen-Based Chelate Employing a "Tethered" Pryidine Sensitizing Moiety Transition To Cyclen-Based Chelates Terbium and Europium Cyclen-Based Chelates for Biomedical Imaging Impurity Produced by the Reaction of Paraformaldehyde and Tributylphosphite Cyclen-Based Chelates with an Overall Net Charge of Zero Principle Energy Lines of the Highly Fluorescent Lanthanides Generalized Jablonski Diagram for the Excitation and Emission of a Europiimi Chelate

3 4 5 7 9 10 11

13 15 17

9. 10. 11.

18 19 20 22 23 26

12. 13.

Absorbance Scan of Terbium and Europium Chelates 3 and 4 Emission Spectra for Chelates 3 and 4

14. Fluorescent Decay Plots 15. Log / versus T Plots 16. Images Resulting from Sryian Hamster Cheek Pouch Model

17. Autofluorescence (excited at 270-31 Onm) and Fluorescence Marker Enhanced (using contrast agent 3) Images 18. 19. Block Diagram of Lifetime Instrument Lifetime Data Acquisition

27 33 34

Vlll

20. 21. 22. 23. 24. 25. 26.

Block Diagram of Fluorescence Imaging Instrument Examples of Light Stable 6-Substituted Quinaldines Model Compounds Employing Three 6-Substituted Quinaldines Model Compounds Chelated with Terbium and Europium Absorbance Scans of the Three Model Compounds Chelated with Terbium Absorbance Scans of the Three Model Compounds Chelated with Europium Emission Spectra of the Model Compound Employing 6-Fluoroquinaldine Complexed with Terbium and Europium Emission Spectra of the Model Compound Employing 6-Chloroquinaldine Complexed with Terbium and Europium Emission Spectra of the Model Compound Employing 6-Methoxyquinaldine Complexed with Europium Cyclen-Based Butyl Half-Ester Employing 6-Fluoroquinaldine Cyclen-Based Butyl Half-Ester Employing 6-Fluoroquinaldine Complexed with Terbium and Europium

35 39 39 46 48 49

50

27.

51

28.

52 66

29. 30.

68 69 70 71

31. 32. 3 3.

Excitation and Emission Spectra of 49 Complexed with Terbium Excitation and Emission Spectra of 49 Complexed with Europium Beer's Law Plot of Unchelated Ligand 49

IX

CHAPTER I INTRODUCTION

Early detection is critical to the clinical outcome in the treatment of cancers (1-3). As an example, colon cancer accounts for \5% of all US cancer related deaths. Once these types of cancer reach metastatic activity there is only a 7% survival rate. In addition, only 37% percent of colon cancers are found early enough for moderate treatment (1). Oral cancer is another example of the necessity for early detection. Each year about 31,000 Americans develop oral cancer. Squamous cell carcinoma (SCC) accounts for 95%) of all malignant oral lesions with SCC having a survival rate of only 50%). Yet when this type of cancer is detected in its earliest stages then the survival rate becomes approximately 80%(4). However, current state-of-the-art detection means often miss early stage colon and oral SCC disease. These detection protocols employ white light endoscopy with gross visualization (5-6). Yet the visual cues for determination of disease state are small, especially the discrimination between non-malignant and dysplastic and pre-malignant lesions. Visual assessment of early lesions within the colon and oral cavities depends on many factors, including the experience of the clinician and his ability to identify the suspect lesions at an early stage of development, and selection of the suspect site that is to be biopsied. Clearly there is a significant need for the enhancement of detection of diseased tissue at its earliest stages to improve the clinical outcome. Alternative techniques to aid in-vivo diagnosis have recently been reported, including the use of visualization techniques to detect changes in the native spectroscopic 1

properties of tissue (7-20). These methodologies are promising, but suffer from an inherently low signal to noise (S/N) ratio and as such are not good candidates for the detection of early lesions. Another factor leading to poor candidacy for detection of early lesions is the fact that absorption and scattering of both excitation and fiuorescence light in the living tissue may induce significant changes in the measured fluorescence signal and render the results of any quantitative measurements rather complicated (21). Because of the limitations associated with detection and diagnoses by native spectroscopic techniques, the use of contrast agents for diagnostic optical imaging has currently experienced an expanded level of attention. The most common compound that has been used clinically is toluidine blue. This compound has been used as a contrast agent for the detection of occult malignancies of the cervix (22); has been found to provide some improvement for non-invasive detection of oral cancer (23-25); and has also been employed in the detection of SCC of the upper aerodigestive tract (26). While the exact mechanism of staining remains unclear, a study of the interaction of toludine blue with tissue using electron microscopy suggests the main factor goveming selective uptake of this dye is the change in cellular membrane permeability (27). It was noted in this report that both injured and malignant lesions exhibit a greater permeability to the dye than that of normal mucosa. Even though toluidine blue does show improvement of sensitivity over conventional white light imaging, it suffers from a lack of a high degree of specificity (23-25). Another example of contrast enhancement agents or site-directed chemical agents that have seen recent success are the PhotoDynamic Therapy (PDT) class of markers (28-29). While these types of markers have shown promise in a diagnostic setting, there are limitations. Long delays for accumulation in tumors.

prolonged photosensitization of skin, and phototoxicity of tissues being imaged are some examples of these limitations (16, 30-33). A need exists for improved contrast agents particularly those that can be used as diagnostic markers in early cancer detection. Recently the use of a pyclen based terbium chelate (Figure 1) has shown promise in detecting chemically induced colon cancers in the Sprague Dawley rat (34). This particular molecule, Tb-[N-(2-pryidylmethyl)-N', N", N"'-tris(methylenephosphonic acid butyl ester)-1,4,7,10 tetraazacyclododecane] or TbPCTMB 1, has excellent fluorescent properties, high specificity, and low toxicity (35-37).

Figure 1 Pyclen-Based Terbium Complex

While currently existing contrast agents in use today suffer from either low specificity, long manifestation times, phototoxicity, or unattractive fluorescent properties this new contrast agent and similar compounds show potential to significantly improve upon these limitations. Some of the spectroscopic properties of Tb-PCTMB that are advantageous to 3

its use as afluorescentcontrast enhancement marker include an extinction coefficient of w3000 L mole' cm' , a high quantum efficiency of 0.51, an extremely large Stokes' shift of 280nm, and a relatively longfluorescentlifetime (2.2ms) (35-37). Figure 2 and Figure 3 shows the absorbance and emission spectmm of Tb-PCTMB 1 respectively. These figures illustrate that the emission signal is spectrally removed from the background, allowing inexpensive instrumentation to be used and low tissue dose levels to be administered. Sensitivity for this class of molecules is such that femtomole/pixel (picomolar) quantities have been quantified in intestinal tissue by endoscopy (35-37).
-1 -1

1.500 A

1 I

270.9 nm 1.370 A

\ I
1

A / \
. 1

1 1

1/

'

1 1 \
1

/ i
'

i.\ i
1

0.700 A

1
\ '

i /

1
1

i i

1
"'-^-....^,'*"'^**'^^'-*'<-vy-HIK^

-0.100 A

190.0 nm

350.0 nm

Figure 2 Absorbance Scan of Tb-PCTMB

300

400

emission wavelength (nm)

500

600

650>

Figure 3 Emission Spectrum of Tb-PCTMB

This high sensitivity allows applications of millimolar solutions to be administered with very low light levels (TLC reader lamp) to be employed during visual detection of suspected sites in the colon of the Sprague Dawley rat (34), with preliminary work indicating that Tb-PCTMB 1 can be used as an exogenous marker for dysplastic tissues with sensitivity as high as 94.7%) (34). While Tb-PCTMB, 1, is an excellentfluorophore,it should be possible to make improvements on the stmcture of the molecule to improve spectroscopic characteristics with the possibility of maintaining its tissue specificity or transport properties. Specifically, red-shifting the absorbance from the far UV into a lower energy, more biologically friendly wavelength region greater than 300nm, would be attractive for several reasons. These include inexpensive optics, easier IR rejection, and potentially less damage to tissues. The goal of this research was to produce lanthanide chelates.

specifically of terbium and europium, which exhibit the attributes necessary to be used as an exogenous markers for early cancer detection, yet with a red-shifted absorbance and improved fluorescence. This was to be accomplished by replacing the pyclen-based molecule with a cyclen-based molecule and then adding a superior light-harvesting moiety that would result in a red-shifted absorption and improved fluorescence. The light-harvesting moiety chosen in this work aimed at improving the spectroscopic properties of the lanthanide chelates is a quinaldine moiety with a functional group in the position six. Use of a quinaldine increases the conjugation compared to a pyridal moiety and therefore is predicted to result in a red-shifted absorbance. Further details about the spectroscopic properties of using a quinaldine moiety as a lanthanide sensitizer is given in Chapter III. Figure 4 gives the stmctures of the target molecules described in this thesis.

0
C4H9O.;' N-

O,
C Tb"^ J
0
;" OC4H9

'

^ P - ^ ^ N Th^^ N

r^0

'-pov_/
PYCLEN BASED
C4H90^g

" 1 *

CYCLEN BASED WITH TETHERED PYRIDYL

^ 0

N N ^ "P/ \ C4H90'^

N^

N; n/ \p^'-' ^"OC4H9

Y: F, CI, OCH3

CYCLEN BASED WITH TETHERED QUINALDINE FOR IMPROVED SPECTROSCOPIC PROPERTIES Figure 4 Target Molecules

CHAPTER II SYNTHESIS AND CHARACTERIZATION OF Tb-[N-(2-PYRIDYLMETHYL)N',N",N"'-TRIS(METHYLENEPHOSPHONIC ACID BUTYL ESTER)-1,4,7,10TETRAAZACYCLODODECANE

2.1 Discussion The use of pyclen based terbium chelates, as previously described above, has shown great promise in the early detection of chemical induced colon cancers in the Sprague Dawley rat. However, from an optical standpoint, the use of an excitation wavelength of 270nm can have disadvantages. Expensive optics, IR rejection, and damage to tissues are some of the major obstacles one incurres when employing these particular derivatives in a diagnostic setting. In order to red-shift the excitation spectrum from the UVA (<300nm) to the UVB region (>300nm), a molecular anterma other than the pyridal moiety must be used (38-41). However, the pyclen molecule is a pryidinebased compound with the antenna incorporated into the chelating cage. It is synthetically more challenging to modify the pyridine moiety to improve the absorption properties and recreate the four-nitrogen macrocycle. Consequently a different macrocycle would have to be used. The use of a cyclen (1,4,7,10-tetraazacyclododecane) -based macrocycle would allow modification of the sensitizing moiety in a more feasible fashion. With prior knowledge of the utility of cyclen compounds and the use of light harvesting moieties (42-53) it was therefore postulated that a cyclen based lanthanide chelate might show both tissue specificity as a cancer contrast agent and have better spectroscopic properties than the pyclen based molecule. The cyclen based molecule with its tethered pyridine, N8

(2-pyridylmethyl)-N', N", N'"-tris(methylenephosphonic acid butyl ester)-1,4,7,10tetraazacyclododecane 2 (figure 5), has previously been described in literature

C4H90"N

''^0C4H9

2
Figure 5 Cyclen-Based Chelate Employing a "Tethered" Pyridine Sensitizing Moiety and is currently patented for use as a MRI contrast enhancement agent (42). This compound, when complexed with gadolinium, has been shown to exhibit low cytoxicity, good water solubility and similar biodistmbution properties as with the pyclen class of molecules (42-43). In addition, derivatives of cyclen based lanthanide chelates for use as fluorescent agents have been reported in the liturature (44-53). So there is evidence to suggest that a cyclen based lanthanide chelate would be a good candidate as a fluorescent contrast agent for diagnostic purposes. Although the exact nature or mechanism of transport which dictates tissue specificity is not known, it was expected that the cyclen-based lanthanide complex would have many of the same properties as the pyclen-based chelate. Net charge, lipophilicity, molecular weight, and overall size are parameters, which are known to influence interstitial and transport properties, and which can be maintained when substituting cyclen-based complexes for pyclen. Thus, it was hypothesized that if these properties can be made to be similar in the cyclen-based complex, then the cyclen analogs might

also display similar tissue specificity to that seen with Tb-PCTMB 1. One way to test this hypothesis would be to attach the pyridal antenna or tether it to a cyclen macrocycle. This approach maintains the four-member nitrogen macrocycle with three phosphobutyl half esters, yet the light harvesting pyridine moiety is "tethered" on to the ring so that it is not an integral part of the macrocycle. Furthermore, from a synthetic standpoint, if the tethered pyridal cyclen lanthanide chelate would show appreciable diagnostic specificity, then the nature of the molecular antenna could more easily be changed to a moiety with red-shifted excitation with the possibility of maintaining its site-directing qualities. As noted in the introduction, the pyclen-based molecule was to be replaced with a cyclenbased molecule and then this cyclen based molecule was to be modified to have redshifted absorption and improved fluorescence. The first step in testing the validity of this transition is shown in Figure 6. Note that in each molecule that the net charge, molecular weight, lipophilicity. and overall size are relatively similar. In the current work, the cyclen-based lanthanide chelate with its tethered pyridal moiety, N-(2-pyridylmethyl)-N', N''. N'"-tris(methylenephosphonic acid butyl ester)-1,4.7,10- tetraazacyclododecane, 2,
0

^ ^ " P ^ N C4H9O /

Tb^^ N N

M^OC4H9 O ^ ^

P^"

,"iJi).rm

C4H9OJ
O -N N N Nij OC4H9

Tb^'

C4H9O ^

PYCLEN BASED Figure 6 Transition to Cyclen-Based Chelates

CYCLEN BASED WITH TETHERED PYRIDYL

10

was synthesized and chelated with terbium and europium. The resulting complexes 3,4 (figure 7) were spectroscopically characterized to examine their utility as a fluorophore. Lastly, the efficacy of the terbium complex 3 was tested through collaboration with Dr. Massoud Motamedi, University of Texas, Medical Branch, Galveston, TX.

0
C4H90.n N N C4H90^ NN-^ J/OC4H9

.rw

3: Ln^^ = Tb^^ 4: Ln^^ = Eu^^ Figure 7 Terbium and Europium Cyclen-Based Chelates for Biomedical Imaging

2.2 Svnthesis 2.2.1 Synthesis of N-(2-pvridvlmethvl)-1,4,7, lO-tetraazacyclododcane 7 The monoakylation of cyclen 5, a species containing four equivalent nitrogens, is accomplished by using a two to one molar ratio of cyclen to the pyridal antennae precursor, 6. Use of high dilution techniques alsofiirthermaximizes the yields (scheme 1). 2-(chloromethyl)pyridine, 6, is commercially available in the protonated form but the free-base form is easily generated in-situ by using two equivalents of potassium carbonate. An alternate way is to generate the free-base form by basifying with an 11

HCl

CI

K2C03

^x^
H \ / H

n
H \ / H

CH.CN

Scheme 1

aqueous hydroxide and extracting into chloroform. Chloroform then becomes the reaction solvent of choice. The polyalkylated impurities were removed by column chromatography employing silica and a chloroform: methanol: ammonium hydroxide solvent system. Yields as high as 75% have been obtained.

2.2.2 Svnthesis of N-(2-pvridvlmethvn-N\ W. N^^^trisCmethylenephosphonicacid butyl ester)-1,4,7,10tetraazacyclododecane 10 The purified pyridine-containing cyclen, 7, was alkylated with methylene phosphonic dibutyl esters. This was accomplished by reacting 7 with paraformaldehyde and tributyl phosphite (scheme 2). The reaction was run under anhydrous conditions and

n
H \ / H

C4H9O.M

1) paraformaldehyde ^ 2) tributylphosphite

C4H9O" C4H90^ C4H90'^

N N

N N-

n
PXOC4H9 O ^ ^

Scheme 2

12

an inert atmosphere using tetrahydrofiiran as the solvent. A common side product was produced by the reaction of paraformaldehyde and tributylphosphite to produce 9 (Figure 8). In order to minimize this side reaction, paraformaldehyde is added first to the reaction vessel containing the monoalkylated 7 and allowed to stir for several hours. The

O JI.OC4H9 HO^ 9 ^OC4H9

Figure 8 Impurity Produced by the Reaction of Paraformaldehyde and Tributylphosphite tributylphosphite was then added slowly followed by stirring until the reaction mixture became clear. Purification included simple distillation to remove the solvent followed by drying under high vacuum to remove the butanol contaminant produced during the reaction. However, when this reaction was scaled up to the multigram level, considerable additional by-products were formed. These impurities were removed by column chromatography using silica and a chloroform: methanol: ammonium hydroxide solvent system to give yields as high as 83%. In order to produce the butyl half ester derivative 10, one butyl group must be removed form each phosphonate moiety. This was accomplished by base hydrolysis, as illustrated in scheme 3. The resulting butyl half ester 10 is now a lipophilic salt with a net charge of negative three. Because the desired ligand now has both organic and water solubility, purification of the resulting potassium hydroxide/product mixture was

13

C4H9O.?

/N=\

C.H.o9

N=

C4H90n

MOC4H9

C4H90^^

^OC4H9

Scheme 3

1^

accomplished by a series of methanol/chloroform washes to separate the excess potassium hydroxidefi-omthe desired product. However care had to be taken with each increase in the chloroform concentration to prevent the reaction of potassium hydroxide and chloroform, which forms dichlorocarbene, a process which is extremely exothermic and the carbene itself is highly reactive towards the final product. First, the reaction solvent was removed under vacuum, producing a darkly colored residue. Then the crude product was taken up in methanol. This forced some of the excess potassium hydroxide to precipitate out. The precipitate was then filtered and the filtrate volume was reduced under vacuum to produce a thick residue. The resulting crude product was then taken up in a 5% chloroform/methanol solution slowly to prevent the production of carbenes. Once again, some of the excess potassium hydroxide was forced out of solution. This whole procedure of filtration, reduction of solvent and addition of chloroform/methanol solution was repeated in increments of increasing 5% chloroform concentrations until 100% is achieved. Once the crude product was taken up in 100% chloroform all the excess potassium hydroxide was removed. The crude product can now be recrystallized in a

14

minimal eimount of chloroform and acetonitrile. The butyl half-ester product 10 was isolated as a white solid with yields ranging from sixty-five to seventy percent.

2.2.3 Complexation with terbium or europium The butyl half-ester derivative 10 was then complexed with either europium or terbium. This was achieved by mixing europium chloride hexahydrate or terbium chloride hexahydrate with the ligand in a 1:1 molar ratio in a slightly acidic aqueous medium. This resulted in a complex with an overall net charge of zero, as illustrated in Figure 9. The salt by-products were then removed by dissolving the complex-salt mixture in 3:1 chloroform: methanol solution andfilteringto remove the precipitated salts.

3: Ln^^ = Tb^^ 4: Ln"^ = Eu"^ Figure 9 Cyclen-Based Chelates with an Overall Net Charge of Zero 15

2.3 Spectroscopic Characterization 2.3.1 Discussion In order to fiilly exploit this cyclen-based class of molecules as fluorescent contrast agents, a complete understanding of their spectral properties is paramount. Excitation, emission spectra, and fluorescent lifetimes are components of information that are necessary in developing proper instrumentation for diagnostic imaging. For example, in order to perform time-resolved fluorescence imaging, information about the fluorescence lifetime decay is critical. By probing for a fluorescence signal at some delay time after excitation, the prompt or autofluorescence and other background signals, seen as scattering, can be eliminated, allowing detection at much greater sensitivity. Therefore, spectroscopically complex systems, such as tissue, can be interrogated giving structural and cellular mechanistic formation and a signal that is temporally shifted from the background. In other words, tissue morphology can be examined based on changes in fluorescence lifetime as a fiinction of tissue pathology (35-37). The process of fluorescence in europium and terbium chelates is attributed to nbond excitation from the So to Si in the organic portion of the ligand. At this point the molecule undergoes a transition to an excited triplet state T. From the triplet state the molecule may either return to the ground state via a spin-forbidden transition (T -> So) or transfer its energy to the ^Di or ^Do orbital of the europium ion or the ^D4 orbital of the terbium ion. This intramolecular energy transfer is unique to lanthanide chelates and can be quite efficient, leading to complexes with high quantum efficiency, (j), values. Figure 10 illustrates the energy of the principle energy lines of the highly fluorescent lanthanides 16

(38). The principle europium ionfluorescenceis through the Do "^ F2 transition while the principle terbium ionfluorescenceis through the ^D4 -> ^Fs transition (38). Due to this uniquefluroescentpathway, luminescent lifetimes are in the millisecond regiem, which is significantly longer than the traditional molecularfluorescentlifetimes ranging from nano to picoseconds. A generalized Jablonski diagram for this entire process of excitation through emission for a europium complex is shown in Figure 11 (38). The complexes prepared under this thesis are expected to follow the same luminescent pathway.

35
6p

7/2

30
'03

o O ^ LU ^ 20
'S/2
'D4 9/2

15

E c en in

c n E c

E c
fi 00

a
c
10

10

F,n/2

If)

V.

0
Sm

6H

5/2

:'F

H 7/2
Gd
3*

'15/2

3*

Eu

Tb

3+

Dy

3*

Figure 10 Principle Energy Lines of the Highly Fluorescent Lanthanides 17

first singlet excited state

triplet donor

acceptor

intersystem crossing

;i^
V C V

mtromolecular energy transfer

u
o 3

thermal decay (HjO)

I
| , I

,,=0
iion fluorescence

I
phosphoresce (ice

o
O

a. E p

excitation

i^

SQ:

i
ligand

i
3+ Eu

'

Figure 11 Generalized Jablonski Diagram for the Excitation and Emission of a Europium Chelate

18

2.3.2 Results The absorption of 3 and 4 is slightly blue-shifted with respect to the pyclen-based molecule with a Xmax of 260nm (Figure 12). The emission spectra for both the terbium and europium complexes are essentially identical to the pyclen-based molecule (Figure 13a-b). As with the pyclen complexes, the principle emission line of terbium (546nm) is red-shifted by approximately 280nm from the excitation while the principle emission line of europium (613nm) is approximately 350nm red-shifted from the excitation.

2.500,

2.OO0-

200.0

22a..a

24O.0

2(0.0

260.0

J 300.0

Figure 12 Absorbance Scan of Terbium and Europium Chelates 3 and 4

19

30000

0
2S000 C4H9O.? N=\
\\ //

?> 20000
C4H90'^ ^ - ' ^^OC4H9

'^

15O00

10000 5000

a
0 ^ 450
'

soo

550

600

650

Wavelengtl: (nm)

<

<sso
EH NL (tics 50/10 na}

7H0>'

Figure 13 Emission Spectra for Chelates 3 and 4

20

Figure 14a and figure 14b shows thefluorescentdecay for the first trial run for the determination of luminescent lifetimes of both the terbium 3 and europium 4 chelates respectively. Data analysis was performed using the SigmaPlot computer software (SPSS Inc.) and the following equations:

/ = (/o)(^"^0 log/ = + log/ r where lo is the luminescent radiant power at the time the excitation source is shut off and T is thefluorescencelifetime decay. A plot of log / versus i gives a straight line with a slope of-l/T. Figure 15a and figure 15b shows the terbium and europium log / versus x plot respectively for the first trial run. Each experiment was repeated four times. The average lifetime determined for the terbium chelate was 3.50ms with a standard deviation of 0.0222. The average lifetime determined for the europium chelate was 0.895ms with a standard deviation of 0.0309.

21

9000 8000

Phosphorescciice Decay of TbTPCTMB at Wavelength 5437 (A) and Temperature 24 C

Trial #1

7000 : 6000 .

I sooo
lH -

ti

4000 3000 2000 1000

J I L J L

1
0.004

Ji

0.000

0.002

0.006

0.008

11me(s)

a
150

Phosphorescence Decay of EuTPCTMB at Wavelength 5939 (A) and Temperature 24 C

Trial #1

100 -
#

50

1
0,0000 0,0005 0.0010 0.0015 0.0020 0.0025 0.0030 0.0035

Time (s)

b
Figure 14 Fluorescent Decay Plots 22

9 50

Phosphorescence Decay of TbTPCTMB at Wavelength 5437 (A) and Temperature 24 C

Trial # 1
t=3.53ms

9,00 S ^ 8.50 8.00 7.50

7.00
^>v

-^^ "^ ^

fi $0

i*vi

0.000

0.002

0,004 Time (s)

0.006

0.008

a
5 00 4.50 ^
tn

Phosphorescence Decay of EuTPCTMB at Wavelength 5939 (A) and Temperature 24 C

Trial #1
t=.909ms

_
_

^?1

4.00

^^'H^ .
- ^ ^ ^ ^ .

1
-5

3.50
3,00 2.50 2.00 1.50 0.0(wo \ 0.0005 .

v^v:

1 0.0010

1 0.0015

. 0.0 320

Time (s)

Figure 15 Log / versus x Plots

23

2.4 Specificity as a Diagnostic Agent 2.4.1 Discussion In collaboration with Massoud Motamedi, UTMB Galveston, TX, the efficacy of using the cyclen-based terbium chelate with its "tethered" pyridal antenna 3 as a cancer contrast agent was tested using a hamster cheek pouch model. The use of a Syrian hamster cheek pouch is a well established model that is the closest model in size and type of tumor that is seen in oral cancer in humans (54). The study consisted of treating Syrian hamster cheek pouches with dimethylbenzanthracene (DMB A) to induce malignant lesions (55) and sodium lauryl sulphate (SLS) to produce nonmalignant lesions (56) and a set of control animals. The following studies were performed using these hamsters: 1) gross inspection of the lesions by white light imaging, 2) autofluorescence of imaging PPIX using excitation light 41 Onm and detecting the changes in fluorescence spectra at 630nm as fimction of lesion development, 3) autofluorescence imaging of elastin, collagen, NAD/NADH and other constituents using excitation light 270-31 Onm without the contrast agent to ensure the emission light collected did not originate from autofluorescence of these common cellular and tissue components, and 4) contrast enhanced images that were collected following topical application of 2 mM solution of the cyclen based terbium chelate with the "tethered" pyridal moiety 3. These studies were performed on control, treated and traumatized cheek pouches. The lesions as detected by autofluorescence, contrast enhancement and gross white light observation were biopsied and H & E staining was performed to assess the diseased state. In addition, the change in fluorescence in a cheek pouch as enhanced by the contrast agent 3 as a ftinction of cancer development from baseline until the lesion was grossly detected 24

was performed. Using these sets of experiments, the efficacy of the contrast agent 3 was determined.

2.4.2 Results Red autofluorescence imaging using excitation at 41 Onm (Figure 16a) shows a lesion that was confirmed upon biopsy and histopathic determination. However, autofluorescence imaging using 270-31 Onm excitation (fig 16b) of the same area did not show lesion development. Upon the application of 3 as a contrast agent, a significantly sharper and clearerfluorescenceimage than that of the autofluorescence, excited at 41 Onm, was observed (Figure 16c). Upon biopsy of this lesion, the histology suggests early lesion development (Figure 16d). It is important to note that this early malignant lesion was not visible to gross white light diagnosis. The white light image of the suspect sites is shovm in fig 16e. The suspect sites as determined by contrast enhancement with 3 are not visible to the naked eye yet were diagnosed as early malignant lesions through standard histology techniques. In addition, no contrast-enhanced images were visible fi*om the cheek pouches of control animals or traumatized cheek pouches using contrast agent 9. Figure 17a-h shows both the autofluorescence (excited at 270-31 Onm) and fluorescence marker enhanced (using contrast agent 3) images of a hamster cheek pouch as a function of malignant lesion development over a nine-week period. Figure 17a-d shows the changes in tissue autofluorescence excited at 270-31 Onm, whereas Figure 17eh shows the changes in contrast enhanced fluorescence. Note that the contrast-enhanced images using 3 provide a clear a marked defined region of early malignant development. 25

Figure 16 Images Resulting from Sryian Hamster Cheek Pouch Model

26

Figure 17 Autofluorescence (excited at 270-31 Onm) and Fluorescence Marker Enhanced (using contrast agent 3) Images

27

2.5 Experimental procedures 2.5.1 General procedure for the synthesis of N-(2-pyridvlmethvlV N \ N^\ N^"- 1.4.7.10-tetraazacyclododecane 7 To a stirring solution of cyclen (2.10g, 0.01219mol) in acetonitrile (310mL) was added high mesh potassium carbonate (1.68g, 0.01219). 2-picolyl chloride hydrochloride (l.OOg, 0.006096mol) was then added slowly to the reaction mixture and allowed to stir until completion. Progress of the reaction was monitored by TLC (silica, 20:4:1 CHCI3: MeOH: NH4OH). After the completion of the reaction the potassium carbonate was removed by filtration through celite and the reaction volume was then reduced to produce a slightly yellow oil. The crude product was then chromatographed on silica with 24:4:1 chloroform: methanol: ammonium hydroxide as the eluent to afford 1.20g (75%) of a shghtly yellow oil that solidified to an off-white solid. H ' N M R (CDCI3): 6 2.3-3.0 (m, 19H), 3.75 (s, 2 H ) , 7.1 (m, IH), 7.45 (d, IH), 7.61-7.70 (m, IH), 8.42-8.51 (m, IH)

2.5.2 General procedure for the synthesis of N-(2-pyridylmethyl)N \ N", N'''-tris(methylenephosphonic acid butyl ester)1,4,7,10-tetraazacyclododecane 10 To stirring solution of N-(2-pyridylmethyl)-N', N", N'"- 1,4,7,10tetraazacyclododecane (Ig, 0.003797mol) in dry THF (30mL) under nitrogen was added paraformaldehyde (378mg, 0.01196mol). The reaction was allowed to stir for three hours at room temperature. Tributylphosphite (3.15g, 0.01196mol) was then added and the mixture was allowed to stir until the solution turned completely clear. The completed reaction mixture was concentrated and dried under high vacuum for 24 hours to afford a pale yellow oil. The oil was then refluxed for four days with aqueous potassium 28

hydroxide (5.75g, 0.1023mol) and enough dioxane to achieve solubility of the starting reactant. The resulting mixture was then reduced in volume under vacuum to produce a thick oil. The oil was then washed with a series of increasing chloroform concentration methanol/chloroform solutions with filtration and removal of solvent. The resulting oil was then dissolved in a minimal amount of chloroform and acetonitrile was then added until the solution became cloudy. The mixture was allowed to stand to precipitate the pure product, which was then filtered, dissolved in water, and lyophilized to produce 2.20g (70%) of a flocculent white solid. H ' NMR (D2O): 6 0.62-0.74 (m, 9H), 0.95-1.11 (m, 6H), 1.13-1.43 (m, 6H), 2.21-3.09 (br m, 20H), 3.49-3.8 (br m, 8H), 7.13-7.18 (m, IH), 7.31-7.37 (d, IH), 7.61-7.69 (m, IH), 8.28-8.34 (m, IH)

2.5.3 General procedure for the chelation of N-(2-pyridylmethynN \ N ' \ N'"-tris(methylenephosphonic acid butyl ester)1,4,7,10-tetraazacyclododecane with terbium 3 The potassium salt of N-(2-pyridylmethyl)-N', N", N"'-tris(methylenephosphonic acid butyl ester)-1,4,7,10-tetraazacyclododecan (300mg, 3.623X10"'*mol) was dissolved in lOOmL of distilled water. The pH of the solution, which was around 10.5 to start, was then adjusted to 5.5 using dilute hydrochloric acid. Terbium (III) chloride hexahydrate (135mg, 3.623X10'^mol) was dissolved in 50mL of distilled water and added to the butyl half-ester solution in one portion with stirring. As the pH began to drop it was maintained around six with a dilute potassium hydroxide solution. Addition of potassium hydroxide was terminated once the pH had settled to around 6.4. The solution was tiien lyophilized, redissolved in a 3:1 chloroform: methanol solution and filtered through celite to remove the resulting potassium chloride salts. The resulting filtrate was then reduced 29

in volume producing a glassy solid. The solid was then taken up in distilled water and filtered through a microfilter to remove Tb(0H)3 and lyophilized to produce a floculant white solid.

2.5.4 General procedure for the determination of absorption and emission wavelengths The absorption spectra for the terbium and europium chelates were acquired with the use of a Shimadzu 265 UV-vis spectrophotometer at room temperature (23C). A 10mm quartz cuvette and a slit width of 1mm were used. The peak find feature was used to identify the maximum relative signal and corresponding wavelength throughout the anaylsis and was used to identify the maximum absorbance wavelength values used in the determination of s. The emission spectra for both the terbium and europium chelates were acquired with an SLM Aminco 4800C fluorimeter at room temperature (23C). A lOmm quartz cuvette and the excitation and emission slit widths were set to 4nim. The excitation wavelength was set to 270nm for the collection of emission spectra.

2.5.5 General procedure for the determination of the fluorescent lifetimes Lifetime measurements of the lanthanide chelates 3, 4 described in this work were made using the spectrometer shown in Figure 18. A low-pressure mercury lamp (Oriel Model C7314) was chosen as the excitation source. All excitation optics incorporated into the optical configuration and the sample cuvettes were comprised of quartz, preventing attenuation of the UV excitation wavelength. A series of lenses were 30

incorporated into the optical configuration to collimate the light from the excitation source (Figure 18, Lens 1) and focus the radiation tightly onto the mechanical chopper blade (EG&G PARC Model 125A) (Figure 18, Lens2). The modulated excitation radiation was then collimated (Figure 18, Lens 3) and focused onto the sample cuvette (Figure 18, Lens 4) housed in an aluminum sample/cuvette holder. The fluorescence emission signal was collected at 90 C with respect to excitation, and focused (Figure 18, Lens 5) onto the entrance slit of the monochrometer (218 GCA McPherson Instruments) set at the emission Am x of the sample [Tb = 543.7 nm (5437 A) and Eu = 613.3 nm (6133 ,a A)]. A PMT (EMI Gencum Inc. Model RFI-5) in communication with an amplifier discriminator (EG&G PARC Model 1120) for noise reduction and a photon counter (EG&G Princeton Applied Research Model 1112) operating in "chopper sample interval" mode in a "piggy back" setup with a boxcar averager (Figure 19) was used for detection. The output signal was then digitized and displayed on a computer (Generic 486 MHz). Chopping frequency, slit width, scan range, scan time, and the photon counter rate vary depending on the organic lanthanide metal complex. First, the chopping frequency for each sample was chosen such that the off-cycle is approximately three times larger than the fluorescence lifetime. A chopping frequency of 170 Hz (3 ms off period) was selected for the Eu-complexed macrocycles and 54 Hz (9.26 ms off period) modulation frequency was chosen for the Tb-chelates. Second, due to the higher fluorescence intensity of the Tb compounds, the entrance and exit slit widths of the monochrometer for the lifetime measurements also varied with respect to Lanthanide metal. Slit widths of 50 pm were selected for the Tb-complexes while larger slit widths of 500 pm were needed 31

to obtain a strong signal for the Eu-chelates. Third, due to the difference in chopping fi-equencies for the Tb and Eu compounds, the scan range of the boxcar averager was varied in accordance with the chopper frequency for each sample. The scan range was set so the gate began coincident with sample excitation. Fourth, the photon counter rate differed for the Tb and Eu-complexes because the Eu fluorescence signal was smaller than the fluorescence of the Tb-samples. For the Tb-complexes, the photon counter was set to take an intensity measurement every 1-second, while a 5-second interval was chosen for the less intense Eu-chelates. Finally, the scan rate of the boxcar was also changed in accordance with the photon counter rate such that a 3-minute boxcar scan was set for the Tb samples and a 8-minute boxcar scan was chosen for the Eu-complex.

2.5.6 General procedure for the determination of biospecificity Malignant lesions in the Syrian hamster cheek were induced by applying topically a 0.5% solution of dimethylbenzanthracene (DMBA) in mineral oil inside the right cheek pouch three times weekly until macroscopic tumors were seen. Induction of visible tumors consistently takes 6-10 weeks. Traumatized cheek pouches with non-malignant lesions were prepared by applying topically sodium lauryl sulphate (SLS) using the same protocol as the DMBA treatment. In a blind study, a pathologist assessed the histology of all lesions. Fluorescence imaging of autofluorescence stemming from PP IX (excitation at 41 Onm, emission at 650nm) were obtained using the fluorescence imaging instrument illustrated in figure 20a while fluorescence from application of contrast agent 3 (excitation at 260nm, emission at 550) were obtained using the fluorescence imaging instrument illustrated in Figure 20b. 32

Hg

Lens#l

Photon Counter

Chopper

Boxcar Averager

Amplifier Discriminator

^^ens # 4 ^ ^

Sample
Lens #5

Figure 18 Block Diagram of Lifetime Instrument

33

Chopper Signal
On period

Off

Boxcar and Photon Counter Signal

Photon Counter Gate

Boxcar Gate Figure 19 Lifetime Data Acquisition

34

Digitizer I.F- 550 nm (5 nm FWHM)

CCD

BP550nm (AA, = 80nm)

UV Lamp @ 260-320 nm

A
Cheek Pouch

Figure 20 Block Diagram of Fluorescence Imaging histrument

35

2.6 Conclusions The results of the early lesion detection employing the Syrian hamster cheek pouch model using Tb-[N-(2-pyridylmethyl)-N', N", N"'-tris(methylenephosphonic acid butyl ester)-1,4,7,10-tetraazacyclododecane] 3 as a fluorescent contrast agent indicate that the cyclen based contrast agent is capable of staining early lesions before they are visible to the naked eye. Only carcinogenic lesions and not lesions originating from traumatized tissue are marked. In addition when compared to PP IX autofluorescence, exciting at 41 Onm with collection of emission at 63 Onm, the cyclen based contrast agent 3 provides a significantly higher contrast for early lesions originating from chemical carcinogens. Furthermore, cyclen-based lanthanide chelates when chelated with either terbium or europium exhibit similar fluorescent properties that make the pyclen-based lanthanide chelate an excellent candidate as a fluorophore for diagnostic imaging. Like the pyclen molecule, the cyclen based chelate with its tethered pyridal moiety exhibits a large Stokes' shift, 280nm for the terbium complex and 355nm for the europium, so the background is spectrally removed from the signal. Also, cyclen based lanthanide chelates exhibit extremely long luminescent lifetimes. The terbium compound 3 has an extremely long fluorescent lifetime of 3.50ms while the europium complex 4 is 0.895ms providing the opportunity to use gated detection to temporally remove the signal from the noise. These results indicate that using a cyclen based lanthanide chelate with a "tethered" antenna as a fluorescent cancer contrast agent shows promise. Thus, opening the avenue of possibly improving the spectroscopic properties of the chelates by employing a quinaldine substituted in position six as a sensitizing antenna while 36

maintaining the tissue specificity properties provides a synthetically much simpler approach than modifying pyclen directly.

37

CHAPTER III SYNTHESIS AND CHARACTERIZATION OF 6-SUBSTITUTED 2.QUIN0LYLMETHYLENEP0LYAZAMACR0CYCL0PH0SPH0NIC ACIDS AND COMPLEXES WITH TERBIUM AND EUROPIUM

3.1 Discussion The results presented in chapter 2 indicate that the cyclen-based terbium complex 3 can be used as a fluorescent contrast agent in the identification of chemically induced early lesions in the Syrian hamster cheek pouch. The next step toward fiirther optimizing the cyclen-based complexes for disease detection involves the modification of the sensitizing moiety, with the hope that this major chemical but minor "structural" modification can be performed while maintaining tissue specificity. Even though the pyclen and cyclen derivatives employing the pryidal moiety require only a moderate photon flux to produce appreciable luminescence to be useful in biomedical imaging (3537) they do have the drawback of requiring a UVA excitation of 270nm. One way to make lanthanide complexes more attractive for biomedical imaging is to employ a molecular antenna with considerable more conjugation in order red shift the excitation. Using a quinoline derivative as the sensitizing moiety at first seemed to represent a way to obtain this red-shifted complex. However, the quinoline molecule and the corresponding quinaldine used to tether to the cyclen macrocycle are photosensitive and decompose when subjected to light. It was then postulated that adding a substituent to the position six of a quinaldine would produce a light stable compound. In addition, there is evidence in the literature to indicate that substituents such as these could improve 38

the quantum yield of the final lanthanide complex (57). Figure 21 shows some examples of 6-substituted quinaldines that are light stable.

8 7c-i^^-^^-^

Y 6^,0-^ Y: F, OCH3, NO2, CF3, CH3

Figure 21 Examples of Light Stable 6-Substituted Quinaldines

In order to examine the utility of the 6-substituted quinaldine derivatives as sensitizing antennas for lanthanides, triphosphonic acid cyclen chelates were synthesized with quinaldine moieties consisting of the fluoro 11, chloro 12, and methoxy 13 derivatives tethered to one of the cyclen nitrogens. These model compounds (Figure 22)

14: Y = F 15:Y = C1 16: Y = OCH3 Figure 22 Model Compounds Employing Three 6-Substituted Quinaldines

39

+3 were then chelated with Tb and Eu^^ and the quantum yield was determined. Once

these complexes were characterized and the 6-fluoro quinaldine was deemed acceptable as a sensitizing moiety antennae for terbium and europium, the more difficult to obtain butyl half-ester derivative was then synthesized.

3.2 Synthesis 3.2.1 Synthesis of 6-Substituted Quinaldines 18-20 The quinaldines synthesized for this study include thefluoro,chloro and methoxy derivatives. The synthesis of this class of compounds was first described by Doebner and von Miller in 1881 (58) and involves reacting para-analines with crotonaldehyde to produce the corresponding quinaldine (scheme 4). However, this synthetic scheme produced many side products and the purification is quite involved.

+
NH2 14:Y = F 15:Y = C1 16:Y = OCH3

H3C

6MHC1 heat
17 18: Y= F 19:Y = C1 20: Y = OCH3 Scheme 4

It was not until 1977 that Charles M. Leir observed that the quinaldine product forms an insoluble sah with ZnCb in acid conditions. This precipitate could then be 40

filtered, washed, decomplexed with ammonium hydroxide and extracted into ether. After removal of solvent and drying, a yellow solid was obtained for all three compounds, with yields in the 50% range (59). The general procedure set forth by Leir was followed in synthesizing the 6-substituted quinaldines. However it was found that the final product could be recrystallized in hexanes to produce a very pure white solid for all three compounds.

3.2.2 Synthesis of 6-Substituted 2(Chloromethyn Quinoline 25-27 In order to tether the quinaldine molecule to the cyclen macrocycle, the quinaldine must befirstfimctionalizedso that attachment to the cyclen is possible. This is done by adding a leaving group to the 2-methyl group of the quinaldine for a typical SN2 reaction with a nitrogen in cyclen. This is achieved by first converting the quinaldine to a quinaldine N-oxide by reaction with 3-chloroperoxybenzoic acid and then a subsequent reaction with tosyl chloride to produce the 6-substituted 2(chloromethyl) quinoline 25-27 (scheme 5).

TsCI CICH2CH2CI Y^Sx^^ ' CICH2CH2CI reflux

18:Y = F 19:Y = C1 20: Y = OCH3

^^^

22:Y = F 23:Y = C1 24: Y = OCH3 Scheme 5

25:Y = F 26:Y = C1 27: Y = OCH3

41

Reaction conditions and purifications schemes employed were similar to those defined by Butera, John A. et. al. (60). In general, the purification after each step included washing with 10% potassium carbonate and ethyl acetate followed by column chromatography using silica. However, it was found that this produced low yields due to high overlap in the column's fractions. Improvement on the purification scheme for the generation of the N-oxide derivative came by mixing the crude product with 10% potassium carbonate and a minimal amount of ethyl acetate and allowing the organic/water bilayer to stand for a couple of days. In this step, the expected product precipitates out. The product was thenfilteredand washed with copious amounts of water to remove potassium carbonate and then dried. Drying was accomplished by dissolving the N-oxide in chloroform followed by the addition of magnesium sulfate, filtration and removal of the organic solvent to produce yields as high as 85%. In the original purification scheme employed for the synthesis of 6-substituted 2(chloromethyl)quinoline 25-27, a silica column produced highly overlapped fractions that resulted in low yields of the desired product. Instead, it was found that a small flash column to remove hexane insoluble by-products followed by recrystalization in hexanes could improve yields by as much as 15%.

3.2.3 Synthesis of N-(6^substituted1-2-quinolylmethyl)l,4,7,10 tetraazacyclododecanes 28-30 The monoalkylation of cyclen, a species containing four equivalent nitrogens, by the quinaldine antennae was achieved by using one equivalent of the quinoline derivative with two equivalents of cyclen. Higher yields can also be achieved by employing high
42

dilution techniques (Scheme 6). Separation of the poly-substituted impurities from the desired monoalkylated product was accomplished through column chromatography

H. r ~ \ , H CHCL H^_J H 25: Y = F 26: Y = CI 27: Y = OCH3 28:Y = F 29: Y = CI 30: Y = OCH3

Scheme 6

using silica and a chloroform: methanol: ammonium hydroxide solvent system to produce a slightly yellow oil that upon standing solidifies to an off white solid. Yields as high as 75% have been obtained.

3.2.4 Synthesis of N-(6[substituted]-2-quinolylmethynN\ N'', N'''-tris(methylene phosphonic acid)1,4,7,10-tetraazacvclododecanes 34-36 Alkylation of the remaining three nitrogens with methylene phosphonic dibutyl esters is necessary to make a functional ligating agent. This product was achieved by first reacting monoalkylated cyclen 28-30 with paraformaldehyde and tributyl phosphite (scheme 7). This reaction was run under anhydrous conditions and an inert atmosphere using dry tetrahydrofuran as the solvent. A common side product 9 was produced by the 43

reaction of paraformaldehyde and tributylphosphite. In order to minimize this side reaction, paraformaldehyde is added first to the reaction vessel containing the monoalkylated cyclen and allowed to stir for several hours. The tributylphosphite was

C4H90.M_^ ^ _ ^

Dparafomialdehyde 2) tributylphoshpite

^^^^^^
U4M9UV C4H90^^

^N L^
/ .

N^ ^J
I \ p^UU4M9 ^"OC4H9

28:Y = F '>c\ ^r ^ 1 29: Y = CI 30:Y = OCH3

31:Y = F Scheme 7 32: Y = CI 33:Y = OCH3

then added slowly followed by stirring until the reaction mixture became clear. Purification usually included simple distillation to remove the solvent followed by drying under high vacuum to remove butanol produced during the reaction. However, when this reaction was scaled up to the multigram level considerable by-products were formed. These impurities were removed by column chromatography using silica and a chloroform: methanol: ammonium hydroxide solvent system to give yields as high as 85%. The removal of the butyl groups to obtain the phosphonic acid derivatives 34-36 was performed by refluxing the dibutyl ester derivatives 31-33 with 6M hydrochloric acid for four days (scheme 8). Purification included azeotropic distillation to remove excess HCl followed by recystalization from anhydrous isopropyl alcohol. The white solid

44

produced by the recystalization was thenfiltered,redissolved in water and then lipholized to produce afloculantwhite solid, with yields as high as 90%.

C4H9O.M C4H90^ C4H90.p_/N^N^p^^OC4H9 C4H9O ^ OC4H9

HO"

?^'
!^"0H

6MHC1 reflux
HO"K

o
31: Y = F 32: Y = C1 33: Y = OCH3 Scheme 8

o
34: Y = F 35: Y = CI 36: Y = OCH3

3.2.5 Complexation with europium and terbium 37-42 The acid derivatives 34-36 were then complexed with either europium or terbium. This was achieved by mixing europium chloride hexahydrate or terbium chloride hexahydrate with the phosphonic acid derivatives in a 1:1 molar ratio in a slightly acidic aqueous medium. This resulted in a complex with an overall charge of negative three (Figure 23).

45

-3

o o o o
37: Ln^^ = Tb^^ Y = F 38: Ln^^ = Tb"^^; Y = CI 39: Ln^^ = Tb^^; Y = OCH3 40: Ln^^ = Eu^^ Y = F 41:Ln^^ = Eu^^Y = Cl 42:Ln^^ = Eu^^;Y = OCH3

Figure 23 Model Compounds Chelated with Terbium and Europium

3.3 Characterization of Lanthanide-N-(6[substituted]-2-quinolylmethylV N', N'', N'''-tris(methylene phosphonic acidV 1,4,7,10 tetraazacyclododecanes 37-42 3.3.1 Discussion Since the ultimate goal of this work is to produce lanthanide chelates with redshifted absorbance, with respect to the pyclen and pyridal cyclen chelates, with tissue specificity, a suitable sensitizing antenna for lanthanides, in particular terbium and europium must first be developed. This sensitizing antenna must not only have a red46

shifted absorbance X,max but also produce sufficient fluorescence in either the terbium (III) or europium (III) ions. In this case, a quinaldine with a substitution in position 6 was chosen for reasons previously stated. Although there are many different quinaldine derivatives available that could represent reasonable candidates, those chosen for this study included the fluoro, chloro, and methoxy derivatives. Triphosphonic acid cyclen derivatives employing a mono-quinaldine antenna 34-36 were chosen as model compounds to test their efficacy as a lanthanide sensitizing moiety due to their simplicity of synthesis. In order to examine the suitability of these new antennae, the excitation, emission, extinction coefficient and quantum yields of the terbium and europium complexes were determined. Since quantitative fluorescence imaging requires an understanding of fluorescence efficiency, it is imperative that the quantum yield be determined. The method employed here for determination of quantum efficiency, (|), parallels established techniques (61-62) and provides an evaluation of the efficiency of photon energy transfer from absorption to fluorescence.

3.3.2 Results The absorbance scans for the three model compounds 34-36 chelated with terbium and europium are given in Figure 24 and Figure 25 respectively. All six compounds show red-shifted absorbance with the amount depending on the type of substituent present on the quinaldine moiety. The measured absorbance maximum are 317, 323, and 330nm for the fluoro, chloro, and methoxy derivatives, respectively. The emission spectra for the six complexes are given in Figures 26-28. No terbium emission for the chelate employing the methoxy quinaldine 39 as an antenna was measured. 47

:ii

.wi

4(X.'

Wavelength (nm)
J
-3
0

1 1
0

NH^/

:P- -X
^p-

Vci

r^/--_//
N N

y=

it

() u -

;; 0

P.'

D
<

s
< v-^
.?0(i r

22,2 320
CO

0.5

4iI

Wavelength (nm)

Wavelength (nm) Figure 24 Absorbance Scans of the Three Model Compounds Chelated with Terbium 48

0)

Xi

o " S ^

^6

ss
U
^t)

31 " J

40V

Wavelength (nm)

11(1

,HM)

4iii I

Wavel engtli (nm)

^(MJ

3'M'

4i III

Wavelength (nm) Figure 25 Absorbance Scans of the Three Model Compounds Chelated with Europium 49

O.OBOO

0-4.
N
N-^

\ =

Tb^O
< -

-0.0001

^<J"
EH M ( t l c a 9 0 / 1 0 nnO L

602

ssoy

0.100

^P^

I ^

.0-

o
I 1

f -0.000

"^ai

>

I.

750>'

E NL ( t i c s 5 0 / 1 0 run) M

Figure 26 Emission Spectra of the Model Compound Employing 6Fluoroquinaldine Complexed with Terbium and Europium

50

<450

500

*-S50

BOO

650>

EH NL (tlCB 80/10 nn)

'830 EM ML (tic* SO/10 n)

Figure 27 Emission Spectra of the Model Compound Employing 6Chloroquinaldine Complexed with Terbium and Europium

51

650 EH HL ( t i c s 50/10 nn]

Figure 28 Emission Spectra of the Model Compound Employing 6Methoxyquinaldine Complexed with Europium

52

The results for the quantum yield determinations of all six model compounds are given in Table 1. The europium chelates for all quinaldine derivatives show increased quantum

Table 1 Lanthanide Chelates, A e , and quantum efficiency <x

Ln+3 Eu+T Tb+3

k ex 318 317
323 322 330

<t)
0.35 0.14 0.50
0.037 0.35

CI CI OCH3 OCH3

Eu+3 Tb+3 Eu+3 Tb+3

329

N/A

-3

yield with respect to the pyclen derivatives of comparable structure. However, the terbium chelates show a decrease in quantum yield with respect to the pyclen derivatives 53

of comparable structure. In addition, the methoxy derivative 39 showed no fluorescence when chelated with terbium.

3.4 Experimental Procedures 3.4.1 General procedure for the synthesis of 6-Fluoroquinaldine 18 4-fluoroaniline (lOg, 0.0900mol) was dissolved in lOOmL of 6 M HCl and heated to a gentie reflux with vigorous stirring. Crotonaldehyde (6.62g, 0.0945mol) was added dropwise to the mixture over a period of 6 hours, after which the reaction was allowed to stir with heat an additional 2 hours. The completed reaction was allowed to cool to room temperature. ZnCb (12.3g, 0.0900mol) was then added to the solution, which was stirred vigorously for 1 hour. The solution was then cooled to 0C and stirred an additional 15 minutes. The precipitate was then vacuumfilteredand washed with chilled 3 M HCl. The precipitate was then transferred to a beaker, stirred with isopropanol for 30 minutes, filtered, washed with additional isopropanol, and thenfinallywith chilled ether. The solid was then transferred to a beaker to which 100 mL water was added and then chilled to 0C while stirring. 30 mL of NH40H(aq) was added to the solution and allowed to stir 10 minutes. The resulting mixture was extracted with chilled ether several times. The combined ether layers were dried using MgS04 and evaporated to afford 7.2g (50% yield) of a slightly yellow solid. The slightiy yellow solid was recrystallized in hexanes to produce a very pure white solid. H ' NMR (CDCI3): 6 2.70 (s, 3H), 7.23-7.45 (m, 3H), 7.94-7.99 (m, 2H).

54

3.4.2 General procedure for the synthesis of 6-fluoroquinaldine N-oxide 22 To a stirring solution of 6-fluoroquinaldine (5g, 0.0310mol) in 1,2 dichloroethane (130ml) was added 3-CPBA (7.43g of 72% activity, 0.0310mol). The reaction was then heated to 40C for 24 hours. The completed reaction mixture was allowed to cool to room temperature, concentrated, and extracted with 10 % K2CO3 and ethyl acetate. A precipitate then formed in both layers and wasfiltered,washed with water to remove traces of K2CO3 and dried to afford 4.67g (85%). H ' NMR (CDCI3): 6 2.66 (s, 3H), 7.287.56 (m, 4H), 8.7-8.79 (qr, IH).

3.4.3 General procedure for the synthesis of 2-(Chloromethyl)6-fluoroquinoline 25 To a stirring solution of p-toluenesulfonyl chloride (6.19g, 0.0325 mol) in dichloroethane (75mL) was added 6-fluoroquinaldine N-oxide (5g, 0.0282mol) under N2. The reaction mixture was then heated to 100C for 24 hours, cooled, concentrated and extracted wdth 10 % K2CO3 and ethyl acetate. The organic layer was dried with MgS04, concentrated and purified on a small silica flash column (2:1 dichloromethane: hexanes). The resulting yellow solid was then recrystallized in hexanes to afford 3.3 Ig (60%) of a white solid. H^ NMR (CDCI3): 6 4.80 (s, 2H), 7.40-7.50 (m, 2H), 7.58-7.61 (d, IH), 8.02-8.99 (m, IH), 8.11-8.14 (d, IH). 3.4.4 General procedure for the synthesis of N-(6-fluoro-2quinolvlmethvlV 1.4,7,10- tetraazacyclododecane 28 To a stirring solution of cyclen (3.52g, 0.0204mol) in chloroform (525mL) was added 2-(Chloromethyl)-6-fluoroquinoline (2g, 0.0102mol). The reaction was then 55

allowed to stir until completion as determined by TLC, concentrated and purified on silica (30:4:1 chloroform: methanol: NH4OH) to afford 2.54g (75%) of a pale yellow oil tiiat solidified on standing to an off-white solid. H* NMR (CDCI3): 6 2.35-3.15 (m, 19H), 3.87 (s, 2H), 7.33-7.42 (m, 2H), 7.58-7.62 (d, IH), 7.94-8.07 (m, 2H).

3.4.5 General procedure for the synthesis of N-(6-fluoro-2quinolylmethylVN\ N^\ N^^'-tris^methvlene phosphonic acid)-1,4.7.10 tetraazacyclododecane 34 To a stirring solution of N-(6-fluoro-2-quinolylmethyl)-l,4,7,10 tetraazacyclododecane (Ig, 0.00302mol) in dry THF (50mL) under N2 was added paraformaldehyde (0.298g, 0.00942mol). The reaction was allowed to stir for 3 hours at room temperature. Tributylphosphite (2.48g, 0.00942mol) was then added to the mixture slowly and allowed to stir until the solution turned completely clear. The completed reaction mixture was concentrated and dried under high vacuum for 24 hours to afford a pale yellow oil. The resulting oil was dissolved in 6 M HCl (50mL) and heated with stirring to a gentle reflux for 4 days. The solution was allowed to cool and excess HCL was removed by azeotropic distillation with water to afford a pale yellow solid. The product was then fiirther purified if necessary by recrystallization with anhydrous isopropyl alcohol to afford 2.17g (90%) of a white solid. The compound was isolated in its fiilly protonated form. H ' NMR (D2O): 6 2.45-3.80 (br m, 22H), 4.07 (s, 2H), 7.677.74 (m, 2H), 7.87-7.91 (d, IH), 8.18-35 (qr IH), 8.79-8.84 (d, IH).

56

3.4.6 General procedure for the svnthesis of 6-chloroquinaldine 19 4-chloroaniline (lOg, 0.0784mol) was dissolved in lOOmL of 6 M HCl and heated to a gentie reflux with vigorous stirring. Crotonaldehyde (5.77g, 0.0823mol) was added dropwise to the mixture over a period of 4 hours after which the reaction was allowed to stir with heat an additional 2 hours. The completed reaction was allowed to cool to room temperature. ZnCb (10.7g, 0.0784mol) was then added to the solution and stirred vigorously for 1 hour. The solution was then cooled to 0C and stirred an additional 15 minutes. The precipitate was then vacuum filtered and washed with chilled 3 M HCl. The precipitate was then transferred to a beaker, stirred with isopropyl alcohol for 30 minutes, filtered, washed with additional isopropyl alcohol, and then finally with chilled ether. The solid was then transferred to a beaker to which 100 mL water was added and then chilled to 0C while stirring. 30 mL of NH40H(aq) was added to the solution and allowed to stir 10 minutes. The resulting mixture was extracted with chilled ether several times. The combined ether layers were dried using MgS04 and evaporated to afford 6.96g (50% yield) of a slightly yellow solid. The slightly yellow solid was then recrystallized in hexanes to produce a very pure white solid. H NMR (CDCI3): 6 2.70 (s, 3H), 7.25-7.27 (d, IH), 7.55-7.59 (m, IH), 7.70-7.71 (d, IH), 7.90-7.92 (d, IH).

3.4.7 General procedure for the synthesis of 6-chloroquinaldine N-oxide 23 To a stirring solution of 6-chloroquinaldine (5g, 0.028 Imol) in 1,2 dichloroethane (130ml) was added 3-CPBA (6.74g with an activity of 72%, 0.0281mol). The reaction was then heated to 40C for 24 hours. The completed reaction was allowed to cool to room temperature, concentrated, and extracted with 10 % K2CO3 and ethyl acetate. A 57

precipitate then fomied in both layers and was filtered and washed with water to remove traces of K2CO3 and dried to afford 4.63g (75%). H ' NMR (CDCI3): 5 2.67 (s, 3H), 7.287.33 (d, IH), 7.49-7.54 (d, IH), 7.60-7.66 (m, IH), 7.77-7.78 (d, IH), 8.66-8.71 (d, IH).

3.4.8 General procedure for the synthesis of 2-rChloromethvlV6chloroquinoline 26 To a stirring solution of p-toluenesulfonyl chloride (5.66g, 0.0297mol) in dichloroethane (75mL) was added 6-chloroquinaldine N-oxide (5g, 0.0258mol) under N2. The reaction mixture was then heated to 100C for 24 hours, cooled, concentrated and extracted with 10 % K2CO3 and ethyl acetate. The organic layer was dried with MgS04, concentrated and purified on a small silica flash column (4:1 dichloromethane: hexanes). The resulting yellow solid was then recrystallized in hexanes to afford 3.0Ig (55%) of a white solid. H ' NMR (CDCI3): 6 4.79 (s, 2H), 7.58-7.70 (m, 2H), 7.77-7.79 (d, IH), 7.95-8.00 (d, IH), 8.07-8.11 (d, IH).

3.4.9 General procedure for the synthesis of N-(6-chloro-2quinolvlmethyl)-1,4,7,10- tetraazacyclododecane 29 To a stirring a stirring solution of cyclen (3.25g, 0.0189mol) in chloroform (525mL) was added 2-(chloromethyl)-6-chloroquinoline (2g, 0.00943mol). The reaction was then allowed to stir until completion, concentrated and purified on silica (30:4:1 chloroform: methanol: NH4OH) to afford 2.48g (75%) of a pale yellow oil that solidified on standing to an off-white solid. 2.42-2.90 (m, 19H), 3.87 (s, 2H), 7.55-7.61 (m, 2H), 7.71-7.72 (d, IH), 7.90-7.93 (d, IH), 8.00-8.05 (d, IH).

58

3.4.10 General procedure for the synthesis of N-(6-chloro-2quinolvlmethvn-N\ N". N'^trisfmethvlene phosphonic acid)-1,4.7,10 tetraazacyclododecane 3^ To a stirring solution of N-(6-chloro-2-quinolylmethyl)-l,4,7,10tetraazacyclododecane (Ig, 0.00287mol) in dry THF (50mL) under N2 was added paraformaldehyde (0.283g, 0.00897mol). The reaction was allowed to stir for 3 hours at room temperature. Tributylphosphite (2.36g, 0.00897mol) was then added to the mixture and allowed to stir until the solution turned completely clear. The completed reaction mixture was concentrated and dried under high vacuum for 24 hours to afford a pale yellow oil. The resulting oil was dissolved in 6 M HCl (50mL) and heated with stirring at a gentle reflux for 4 days. The solution was allowed to cool and excess HCL was removed by azeotropic distillation with water to afford a pale yellow solid. The product was then fiirther purified if necessary by recrystallization with isopropyl alcohol to afford 2.03g (87%) of a white solid. The compound is isolated in its fiilly protonated form. H^ NMR (D2O): 5 2.55-3.90 (br m, 22H), 4.15 (s, 2H), 7.38-7.41 (d, IH), 7.51-7.60 (m, IH), 7.80-7.89 (d, IH), 8.07-8.21 (d, IH), 8.70-8.79 (d, IH).

3.4.11 General procedure for the synthesis of 6-methoxyquinaldine 20 /7-Anisidine (lOg, 0.0812mol) was dissolved in lOOmL of 6 M HCl and heated to a gentie reflux with vigorous stirring. Crotonaldehyde (5.98g, 0.0853mol) was added dropwise to the mixture over a period of 4 hours after which the reaction was allowed to stir with heat an additional 2 hours. The completed reaction was allowed to cool to room temperature. ZnCb (11.07g, 0.0812mol) was then added to the solution and stin-ed vigorously for 1 hour. The solution was then cooled to 0C and stin-ed an additional 15 59

minutes. The precipitate was then vacuum filtered and washed with chilled 3 M HCl. The precipitate was then transfen-ed to a beaker, stin-ed with isopropyl alcohol for 30 minutes, filtered, washed with additional isopropyl alcohol, and then finally with chilled ether. The solid was then transfened to a beaker to which 100 mL water was added and then chilled to 0C while stining. 30 mL of NH40H(aq) was added to the solution and allowed to stir 10 minutes. The resulting mixture was extracted with chilled ether several times. The combined ether layers were dried using MgS04 and evaporated to afford 5.07g (37% yield) of a slightly yellow solid. The slightiy yellow solid was recrystallized in hexanes to produce a very pure white solid. H ' NMR (CDCI3): 5 2.68 (s, 3H), 3.89 (s, 3H), 7.01-7.02 (d, IH), 7.22-7.24 (d, IH), 7.28-7.32 (m, IH), 7.87-7.92 (m, 2H).

3.4.12 General procedure for the synthesis of 6-methoxyquinaldine N-oxide 24 To a stirring solution of 6-methoxyquinaldine (5g, 0.0289mol) in 1,2 dichloroethane (130ml) was added 3-CPBA (6.93g with an activity of 72%, 0.0289mol). The reaction was then heated to 40C for 24 hours. The completed reaction was allowed to cool to room temperature, concentrated, and extracted with 10 % K2CO3 and ethyl acetate. A precipitate then formed in both layers and was filtered and washed with water to remove K2CO3 and dried to afford 4.65g (85%). H ' NMR (CDCI3): 6 2.64 (s, 3H), 3.89 (s, 3H), 7.03-7.05 (d, IH), 7.21-7.24 (d, IH), 7.30-7.37 (m, IH), 7.48-7.52 (d, IH), 8.63-8.66 (d, IH).

60

3.4.13 General procedure for the svnthesis of 2-rChloromethyn6-methoxyquinoline 27 To a stirring solution of p-toluenesulfonyl chloride (5.79g, 0.0304mol) in dichloroethane (75mL) was added 6-methoxyquinaldine N-oxide (5g, 0.0264mol) under N2. The reaction mixture was then heated to 100C for 24 hours, cooled, concentrated and extracted with 10 % K2CO3 and ethyl acetate. The organic layer was dried with MgS04, concentrated and purified on silica (1:1 dichloromethane: hexanes) to afford 1.92g (35%) of a white solid. H ' NMR (CDCI3): 6 3.91 (s, 3H), 4.79 (s, 2H), 7.04-7.06 (d IH), 7.34-7.38 (m, IH), 7.51-7.54 (d, IH), 7.92-7.95 (d, IH), 8.04-8.08 (d IH).

3.4.14 General procedure for the synthesis of N-(6-methoxy2-quinolylmethyl)-1,4,7,10 tetraazacyclododecane 30 To a stirring a stirring solution of cyclen (3.32g, 0.0193mol) in chloroform (525mL) was added 2-(Chloromethyl)-6-methoxyquinoline (2g, 0.00963mol). The reaction was then allowed to stir until completion, concentrated and purified on silica (30:4:1 chloroform: methanol: NH4OH) to afford 2.55g (77%) of a pale yellow oil that solidified on standing to an off-white solid. H ' NMR (CDCI3): 6 2.31-2.95 (m, 19H), 3.86 (s, 2H), 3.88 (s, 3H), 7.00-7.01 (d, IH), 7.26-7.31 (m, IH), 7.52-7.55 (d, IH), 7.877.90 (d,lH), 7.99-8.01 (d,lH).

3.4.15 General procedure for the synthesis of N-(6-methoxy2-quinolylmethyn-N\ N". N'^^-tris(methylene phosphonic acid)-1,4,7.10 tetraazacyclododecane 36 To a stin-ing solution of N-(6-methoxy-2-quinolylmethyl)-l,4,7,10 tetraazacyclododecane (Ig, 0.0029Imol) in dry THF (50mL) under N2 was added 61

parafonnaldehyde (0.287g 0.00908mol). The reaction was allowed to stir for 3 hours at room temperature. Tributylphosphite (2.39g, 0.00908mol) was then added to the mixture and allowed to stir until the solution tumed completely clear. The completed reaction mixture was concentrated and dried under high vacuum for 24 hours to afford a pale yellow oil. The resulting oil was dissolved in 6 M HCl (50mL) and heated with stining to a gentie reflux for 4 days. The solution was allowed to cool and excess HCL was removed by azeotropic distillation with water to afford a pale yellow solid. The product was then fiirther purified if necessary by recrystallization with isopropyl alcohol to afford 2.07g (88%) of a white solid. H ' NMR (D2O): 6 2.50-3.81 (br m, 22H), 4.09-4.25 (br m, 5H), 7.13-7.16 (d, IH), 7.26-7.32 (m, IH), 7.40-7.43 (d, IH), 7.51-7.54 (d IH), 8.65-8.69 (d, IH).

3.4.16 General procedure for the chelation of the N-(6rsubstituted12-quinolylmethyl)-N\ N'\ N'''-tris(methylene phosphonic acid)1.4,7,10 tetraazacyclododecanes with terbium or europium 37-42 Using N-(6-fluoro-2-quinolylmethyl)-N', N", N'"-tris(methylene phosphonic acid)-1,4,7,10 tetraazacyclododecane and europium chloride hexahydrate as an example: to N-(6-fluoro-2-quinolylmethyl)-N', N", N"'-tris(methylene phosphonic acid)-1,4,7,10 tetraazacyclododecane (300mg, 3.770X10"'*mol) was added distilled water (lOOmL). The pH of the solution was adjusted to six using dilute KOH solution. Europium chloride hexahydrate (128mg, 3.77X10"'^mol) dissolved in distilled water (50mL) was then added slowly to the solution with stirring. The pH dropped and was maintained at 6 using the dilute potassium hydroxide solution. After the complete addition of the europium solution the mixture was allowed to stir until a pH of 6 was maintained. The resulting 62

chelate solution was then lypholized, dissolved in lOOmL of distilled water, filtered tiirough a micro filter to remove Eu(0H)3, and lypholized again to produce a white floculant solid.

3.4.17 General procedure for the determination of quantum efficiency The quantum efficiency, ^, for the six complexes 37-42 was determined by using the following equation:

o =0
s

A/^.-2\ f?>

v^.y \^r

(I-IQ-'^^O -OD. (I-IQ-'^^O JL

-OD,

where D is the integrated area of the corrected fiuorescence curve in arbitrary units, 7 is 7 the refractive index, OD is the optical density in arbitrary units. The subscripts s and r are the sample and reference respectively. Rhodamine 6G was used as the reference. The absorption spectra for the terbium and europium chelates 37-42 were acquired with the use of a Shimadzu 265 UV-vis spectrophotometer at room temperature (23 C). A 10mm quartz cuvette and a slit width of 1mm were used. The peak find feature was used to identify the maximum relative signal and corresponding >.max throughout the anaylsis. The maximum absorbance wavelength values used were the farthest red-shifted maximum for the complexes. The emission spectra for both the terbium and europium chelates 37-42 were acquired with an SLM Aminco 4800C fluorimeter at room temperature (23C). A 10mm quartz cuvette and the excitation and emission slit widths were set to 4mm. The

63

excitation wavelength was set to corresponding A^^ax for each of the complexes and the reference, rhodamine 6G, for the collection of emission spectra. Refractive index values for all six complexes 37-42 and the rhodamine 6g standard solution were measured with a Bausch & Lomb refractive index instrument (Cat. No. 33-45-59). All refractive index values were obtained at 25C.

3.5 Conclusions The three model compounds synthesized employing the fluoro 34, chloro 35, and methoxy 36 derivatives of quinaldine have red-shifted absorbance with respect to the pyclen and cyclen based complexes employing the pyridal moiety with an absorbance of 270nm and 260nm, respectively. In addition the nature of the 6-substituent dictates how far the absorbance will be shifted. With a >.max of 317, 323, and 330nm for the fluoro, chloro, and methoxy derivatives, respectively, the possibility exist for additional substituents to red-shift the absorbance even fiirther. For example, a nitro or a trifluoromethyl moiety might move the excitation even fiirther and improve the quantum efficiency. The europium complexes have (t)-values of 0.35, 0.50, and 0.35 for the fluoro, chloro, and methoxy quinaldine derivatives, respectively. For comparison, the Eu^^pyclen triphosphonic acid complex has a (|)-value of 0.16. This improvement in quantum efficiency is a 119% increase for the fluoro and methoxy derivatives while the chloro represent a 212%> increase. The europium model compound values are a significant improvement with respect to the pyclen complexes. The terbium complexes have ^values of 0.14, and 0.037 for the fluoro and chloro quinaldine derivatives. The methoxy quinaldine derivative did not have significant terbium emission. The terbium pyclen 64

triphosphonic acid complex has a (t)-value of 0.30. These values show a decrease in quantum efficiency with respect to the pyclen complexes.

65

CHAPTER IV SYNTHESIS OF N-(6-FLUORO-2-QUINOLYLMETHYL)- N', N", N' TRIS(METHYLENE PHOSPHONIC ACID BUTYL ESTER)1,4,7,10-TETRAZAC YCLODODEC ANE AND COMPLEXES WITH TERBIUM AND EUROPIUM

4.1 Introduction Spectroscopic characterization of the triphosphonic acid model compounds 37-42 demonstrated that some quinaldine derivatives are sufficient sensitizers for both europium and terbium. All three quinaldine derivatives 40-42 synthesized proved to be sufficient sensitizers for europium with ^ values of 0.34 for thefluoroand methoxy derivatives and 0.50 for the chloro derivative. Thefluoroderivative 37 demonstrated sensitization for terbium with a (j) value of 0.14. However with the chloro derivative 38 had only moderate sensitization for terbium with a (j) value 0.037 and terbium sensitization was not detected with the methoxy derivative 39. It was therefore decided that 6-fluoroquinaldine should be employed as the sensitizing moiety in the more difficult to obtain yet desired butyl half-ester derivative (Figure 29).
P K O ^N N-^ N \ ^
^ ^

K^-0.p_3NN^p.0-K^
C4H90^^ ^"OC4H9

49
Figure 29 Cyclen-Based Butyl Half-Ester Employing 6-Fluoroquinaldine 66

4.2 Synthesis 4.2.1 Synthesis of N-r6-fluoro-2-quinolylmethvl)-N\ K'. ^"'trisfmethvlene phosphonic acid butyl esterVl ,4,7J 0tetraazacvclododecane 43 In order to produce the butyl half-ester derivative, just one butyl group must be removed from each of the phosphonic ester groups. This was accomplished by base hydrolysis (scheme 9). In the case of base hydrolysis, as opposed to acid hydrolysis, when one butyl group is removed this produces a negative charge on the phospho moiety that prevents attack of another negatively charged hydroxyl group, thereby producing the

C4H9O.M C4H9O'

^ fg ^ N ^

V = /

C4H9O.-.

_ / /

\ ^ /

C4H90.p_/N C4H90'M

OC4H9 n-OC4H9

4 days

OVp_:;N C4H90^^ ^

N^^^O" ^OC4H9

Scheme 9

butyl half-ester derivative. However in acid hydrolysis, when one butyl group is removed a neutrally charged phoshpo moiety is produced which does not prevent attack from another positively charged hydronium ion, thereby producing the fiill acid derivative. The resulting butyl half ester 43 is now a lipophilic salt with a net charge of negative three. Purification was accomplished in the same manner as described for N-(2pyridylmethyl)-N', N", N"'-tris(methylenephosphonic acid butyl ester)-1,4,7,10tetraazacyclododecane 10 described in Chapter II, section 2.2.

67

4.2.2 Complexation with terbium and europium The general procedure which was used for complexing the butyl half-ester derivative 43 is the same as for the acid derivatives 34-36 described in 3.2.5. However in this case the resulting molecule has a net charge of zero (Figure 30).

o
C4H90.n

44: Ln^3 = Tb^^ 45: Ln^^ = Eu^^ Fig 30 Cyclen-Based Butyl Half-Ester Employing 6-Fluoroquinaldine Complexed with Terbium and Europium

4.3 Spectroscopic Characterization The emission and excitation scans for the terbium and europium complexes with the butyl half-ester derivative 44,45 are given in Figure 31 and Figure 32. Like the pyclen complexes and the model compounds, the emission is far removed from the background noise. The extinction coefficient was determined for the free ligand 43 at its ^max of 317nm. This determination was done by making a series of solutions of known concentration diluted by one-half, obtaining the absorbance value at 317nm, and ploting 68

MtWWWMWfc.* I

lA. Hitft m/i9 MB

.-->^,-y^ aifetfr

_400J

0
C4H90.^

0"

y'
^H

1
^C,H,0 ^^

Lib"'

1 OC4H 9 1

1.000

L /.< \

: Jv.__^^ L ^ ^ ^^:^^-/^^^^-^^
twtya*

Figure 31 Excitation and Emission Spectra of 49 Complexed with Terbium

69

r1.000

a
250 300 350 4C0

0
C4H90.M

-xr~A
^N ^N

N N

Eu

ov

C4H90 ^

p - -^ \_y

11.000

:i I
t-

^
.e40

III
700
:!tj

400

500

f300

Figure 32 Excitation and Emission Spectra of 49 Complexed with Europium

70

absorbance vs. concentration. The resulting plot produces a straight line with the slope of 8 according to Beer's Law

A=8bc

where A is the absorbance, 8 is the extinction coefficient, b is the pathlength of the cuvette used in the absorbance measurement and c is the concentration of the solution. An extinction coefficient value of 4676L/mol*cm was calculated (Figure 33).

D.25 slope 467B.BB53D20294 r* 0.9923568548 Y-int 0.0308985285 0.20

D.15

E
in

G.10

0.05

OOO

1e-5

2e-5

3e-5

4e-5

5e-5

Concentration

Figure 33 Beer's Law Plot of Unchelated Ligand 49

71

4.4 Experimental Procedures 4.4.1 General procedure for the svnthesis of N-r6-fluoro-2quinolylmethyn-N\ N^'. N^^^tris(methvlene phosphonic acidbutvl esterVl.4.7.10 tetraazacyclododecane 43 To a stirring solution of N-(6-fluoro-2-quinolylmethyl)-l,4,7,10tetraazacyclododecane (Ig, 0.00302mol) in dry THF (50mL) under N2 was added paraformaldehyde (0.298g, 0.00942mol). The reaction was allowed to stir for 3 hours at room temperature. Tributylphosphite (2.48g, 0.00942mol) was then added to the mixture and allowed to stir until the solution tumed completely clear. The completed reaction mixture was concentrated and dried under high vacuum for 24 hours to afford a pale yellow oil. The oil was then refluxed for four days with 27 equivalents of KOH dissolved in 20mL of H2O with enough dioxane to achieve solubility. The resulting mixture volume was then reduced under vacuum to produce a thick oil. The oil was then washed with a series of increasing chloroform concentration methanol/chloroform solutions with filtration and removal of solvent. The resulting oil was then dissolved in a minimal amount of chloroform and acetonitrile was then added until the solution became cloudy. The mixture was allowed to stand to precipitate the pure product which was then filtered, dissolved in water, and lypholized to produce 1.75g (65%) of a white, flaky solid. H^ NMR (D2O): 6 0.75-0.85 (m, 9H), 1.18-1.45 (m, 6H), 1.48-1.55 (m, 6H), 1.913.10 (brm, 16H), 3.65-3.82 (brm, 12H), 4.10-4.21 (br, 2H), 7.32-7.41 (m, IH), 7.50-7.61 (br, IH), 7.80-7.88 (d, IH), 7.92-8.01 (m, IH), 8.20-8.31 (d, IH).

72

4.4.2 General procedure for the chelation of N-r6-fluoro-2quinolylmethyn-N\ N". N'^trisrmethvlene phosphonic acid butyl esterV 1.4.7.10 tetraazacyclododecane with terbium or europium 44-45 The potassium sah of N-(6-fluoro-2-quinolylmethyl)-N', N", N'"-tris(methylene phosphonic acid butyl ester)-1,4,7,10 tetraazacyclododecane (300mg, 3.34X10"^mol) was dissolved in lOOmL of distilled water. The pH of the solution, which was around 10.5 to start, was then adjusted to 5.5 using dilute hydrochloric acid. Europium chloride hexahydrate (123mg, 3.34X10'^mol) was dissolved in 50mL of distilled water and added to the butyl half ester solution in one portion with stirring. As the pH began to drop, it was maintained around six with a dilute potassium hydroxide solution. Addition of potassium hydroxide was terminated once the pH had settled around 6.4. The solution was then lypholized, redissolved in a 3:1 chloroform: methanol solution and filtered through celite. The resulting filtrate was then concentrated producing a glassy solid. The solid was then taken up in water and filtered through a microfilter to remove Eu(0H)3 and lipholized to produce a floculant white solid.

4.4.3 General procedure for the determination of absorption and emission wavelengths The absorption spectra for the terbium and europium chelates were acquired with the use of a Shimadzu 265 UV-vis spectrophotometer at room temperature (23C). A 10mm quartz cuvette and a slit width of 1mm were used. The peak find feature was used to identify the maximum relative signal and corresponding wavelength throughout the anaylsis and was used to identify the maximum absorbance wavelength values used in the determination of 8. 73

The emission spectra for both the terbium and europium chelates were acquired with an SLM Aminco 4800C fluorimeter at room temperature (23C). A 10mm quartz cuvette and the excitation and emission slit widths were set to 4mm. The excitation wavelength was set to 317nm for the collection of emission spectra.

4.4.4 General procedure for the determination of molar extinction coefficient The butyl half ester 43 was dissolved in deionized water to produce a solution of concentration 2 X 10"^M. The solution was then diluted 1:2 sequentially to make four additional solutions that were subsequently used as standards in the determination of s according to the Beer-Lambert relationship. The maximum absorbance values were obtained using the same UV-vis spectrophotometer as described in chapter 3 section 19. These values were then plotted against the corresponding concentration to produce a linear line with a slope equal to 8 according to the following equation:

A=sbc
where A is the absorbance, c is the concentration, b is the pathlength and s is the molar extinction coefficient.

4.5 Conclusions The butyl half ester derivatives employing 6-fluoroqiunaldine 43 chelated with terbium (compound 44) and europium (compound 45) as lanthanide sensitizers have been successfully synthesized. The resulting emission for the europium complex has more features than has been previously seen with the pyclen or cyclen based compounds. The 74

reason for this apparent spectral line narrowing is under investigation. Further study is required to explain these structural effects on the emission of lanthanide chelates employing sensitizing moieties. In addition, expanded spectroscopic characterization, including quantum yield and the extinction coefficient, needs to be determined. Preliminary cytoxicity studies are currently underway to determine if its use as a fluorescence contrast agent in the detection of early malignant lesions is feasible.

75

CHAPTER V CONCLUSIONS AND FUTURE PROJECTS

Preliminary results suggest that using cyclen-based lanthanide chelates, as opposed to pyclen-based, as fluorescent contrast agents to detect early lesions associated v^th oral squamous cell carcinoma is promising. Although the exact mechanism of transport or tissue specificity is unknown, the cyclen-based complexes do have many of the same properties of the pyclen-based molecules. By keeping the relative molecular weight, net charge, lipophilicity, and overall size similar to the pyclen-based molecule it was demonstrated that the cyclen based triphosphonic acid butyl half ester with the tethered pyridine 3 displayed similar tissue specificity as does the pyclen molecule. It was shown in Chapter III that 6-fluoroquinaldine provides adequate sensitization for both europium and terbium. Modification of the pyridal moiety to a quinaldine moiety should not affect the overall size, lipophilicity, net charge and relative molecular weight. Therefore the more difficult-to-obtain butyl half-ester employing 6fluoroquinaldine 27 as a sensitizing moiety was synthesized and exhibits luminescence when chelated with either terbium or europium. Expanded spectroscopic characterization and cytoxicity studies are currently being investigated. These studies are paramount in order to fully utilize this new lanthanide chelate as a fluorescent contrast agent. The quantum yield and luminescent lifetimes of both the terbium and europium complexes are components of information that are necessary in developing proper instrumentation for diagnostic imaging. In

76

addition this newfluorescentcontrast agent must not be cytotoxic for applications in biological systems. Additional quinoline derivatives are being investigated as suitable sensitizing moieties for terbium and europium. Cyclen based chelates employing 6-nitroquinaldine, 6-trifluoromethylquinaldine and isoquinoline are currently being planned to be synthesized as model compounds to test their efficiency of sensitization towards terbium and europium.

77

LITERATURE CITED 1. Cancer Facts and Figures, American Cancer Society, Atianta, 1997, 17. 2. W.F. Enneking, E.U. Conrad, "Common bone tumors," Clinical Symposia, CibaGeiegy, 1989, 4i, 3-32. 3. M. Chackal, B.R. Zetter, Proc. Natl. Acad USA, 1992, 89, 6197-6201. 4. S. Silverman, M. Gorsky, "Epidemiologogic and demographic update in oral cancer: California and national data1973 to 1985," J Am. Dent. Assoc, 1990, 120, (5), 495-499. 5. P. B. Cotton, Pratical Gastrointestinal Endoscopy, Oxford, Science, 3'"* Ed., 1990. 6. D. M. Melville, J. R. Jass, B. Morson, D. J. Pollock, P. I. Richman, N. A. Shepard, J. K. Ritchie, S. B. Love, J. E. Lennard-Johnes, "Observer study of the grading of dysplasia in ulcerative colitis: comparison with clinical outcome," Human Pathology, 1989,20, 1008. 7. J. Leonard, W. Beck, "Hematoprophyrin fluorescence: an aid in diagnosis of malignant neoplasms," Laryngoscope, 1971, 81, 365-372. 8. A. Gillenwater, R. Jacob, R. Richards-Kortum, "Fluorescence spectroscopy: a technique with potential to improve the early detection of aerodigestive tract neoplasia," Head and Neck, 1998, 20, (6), 556-562. 9. D. Harris, J. Hill, J. Werkhaven, E. Applebaum, R. Lobraico, W. Waldo, "Porphyrin fluorescence and photosensitization in head and neck cancer," Otolaryngol Head Neck Surg, 1986,112, 1194-1199. 10. R. M. Cothren, M. V. Sivak Jr., J. Van Dam, R. E. Petras, M. Fitzmaurice, J. M. Crawford, J. Wu, J. F. Brennan, R. P. Rava, R. Manoharan, M. S. Feld, "Detection of dysplasia at colonoscopy using laser-induced fluorescence: a blinded study," Gastrointest. Endosc, 1996, 44, (2), 168-176. 11. D. M. Harris, J. Werkhaven, "Endogenous porphyrin fluorescence in tumors," Lasers Surg Med, 1987, 7, (6), 467-472. 12. J. Hung, S. Lam, B. Placic, "Autofluorescence of normal and malignant bronchial tissue," Lasers Surg Med, 1991, H , 99-105.

78

13. S. Lam, T. Kennedy, M. Unger, Y. Miller, D. Gelmont, V. Rusch, B. Gipe, D. Howard, J. LeRiche, A. Gazdar, "Localization of bronchial intraepithelial neoplastic lesions by fluorescence bronchoscopy," Chest, 1998,113, (3), 696-702. 14. R. Richards-Kortum, R. P. Rava, R. E. Petras, M. Fitzmaurice, M. Sivak, M. S. Feld, "Spectroscopic diagnosis of colonic Dysplasia," Photochem. Photobiol, 1991, 53, (6), 777-786. ^ ^^ 15. H. J. Sterenborg, S. Thomsen, S. L. Jacques, M. Duvic, M. Motamedi, R. F. Wagner Jr., "In-Vivo fluorescence spectroscopy and imaging of human skin tumors," [letter], Dermatologic Surgery, 1995, 21, (9), 821-822. 16. H. J. C. M. Sterenborg, M. Motamedi, R. F. Wagner, M. Duvic, S. Thomsen, S. L. Jacques, "In-Vivo fluorescence spectroscopy and imaging of human skin tumors," Lasers Surg. Med, 1994,9, 191-201. 17. S. Abe, K. Izuishi, H. Tajiri, T. Kinoshita, T. Matsuoka, "Original ArticleCorrelation of In-Vitro Autofluorescence Endoscopy Images with Histopathologic Finding in Stomach Cancer," Endoscopy, 2000, 32, (4), 281. 18. M. Zargi, I. Fajdiga, L. Smid, "Autofluorescence Imaging in the Diagnosis of Laryngeal Cancer," European archives ofoto-rhino-laryngology: official journal of the European Federation of Oto-Rhino-Laryngological Societies (EUFOS), 2000, 257,(1), 17. 19. M. Inaguma, K. Hashimoto, "Porphrin-like Fluorescence in Oral Cancer: In-Vivo Fluorescence Spectral Characteristics of Lesions by Use of a Near-Ultraviolet Excited Autofluorescence Diagnosis System and Separation of Fluorescent Extract by Capillary Electrophoresis," Cancer, 1999, 86, (11), 2201. 20. C. S. Betz, M. Mehlmann, K. Rick, H. Stepp, G. Grevers, R. Baumgartner, A. Leung, "Autofluorescence Imaging and Spectroscopy of Normal and Malignant Mucosa in Patients with Head and Neck Cancer," Lasers Surg. Med., 1999, 25, (4), 323. 21. K. Schomacker, J. Frisoli, C. Compton, T. Flotte, J. Richter, T. Deutsch, N. Nishioka, "Ultraviolet laser-induced fluorescence of colonic polyps," Gastroenterology, 1992, 102.1155-1160. 22. R. Richart, "A clinical staining test for the in-vivo delineation of Dysplasia and carcinoma in-situ," .4m. J. Obstet. Gyneol, 1963, 86, 703. 23. J. Little, "Toludine blue staining in the detection of oral precancerous and malignant lesions," Oral Medicine, 1984, 57, 379.

79

24. G. Miller, W. Maurer, M. Savary, P. Monnier, F. Gloor, "A case of oesophageal cancer limited to the mucosa and submucosa," Endoscopy, 1979, U , (3), 175-178. 25. J. B. Epstein, C. Scully, J. Spinelli, "Toluidine blue and Lugol's iodine application in the assessment of oral malignant disease and lesions at risk of malignancy," J Oral Pathol. Med, 1992, 21, (4), 160-163. 26. W. Hix, "Toluidine blue staining of the esophagus," y4rc/z. Otolaryngol. Head Neck S-wrg., 1987,113, 864. 27. P. Herlin, "A study of the mechanism of the toludine blue dye test," Endoscopy, 1983,15, 4. 28. R. H. Pottier, Y. F. A. Chow, J. LaPlante, T. Truscott, J. C. Kennedy, L. A. Beiner, "Non-invasive technique for obtaining fluorescence excitation and emission spectra in-vivo," Photochemistry and Photobiology, 1986, 44, (5), 679-687. 29. T. Mang, C. McGinnis, C. Liewbow, U. Nseyo, D. Crean, T. Dougherty, "Fluorescence detection of tumors, early diagnosis of microscopic lesions in preclinical studies," Cancer, 1993, 71, 269-276. 30. J. Kennedy, R. Pottier, "Endogenous protoporhyrin IX, a clinically useful photosensitizer for photodynamic therapy," J. Photochem. Photobiol., 1992,14, 275292. 31. W. E. Grant, C. Hopper, P. M. Speight, A. J. MacRobert, S. G. Bown, "Photodynamic therapy of malignant and premalignant lesions in patients with 'field cancerization' of the oral cavity," J Laryngol. Otol, 1993,107, (12), 1140-1145. 32. R. Szeimies, T. Sassy, M. Landthaler, "Penetration potency of topical applied Aminolevulinic acid for photodynamic therapy of basal cell carcinoma," J. Photochem. Photobiol, 1994, 59, (1), 73-76. 33. A. Leunig, C. S. Betz, M. Mehlmann, S. Stepp, S. Arbogast, G. Grever, R. Baumgartner, "Detection of squamous cell carcinoma of the oral cavit by imaging 5aminolevulinic acid-induced protoporphyrin IX fluorescence," Lryngoscope, 2000, 110,(1), 78-83. 34. M. P. Houlne, D. S. Hubbard, G. E. Kiefer, D. J. Bomhop, Texas Tech University, unpublished data 35. D. S. Hubbard, M. P. Houlne, G. Kiefer, H. F. Janssen, C. Hacker, D. J. Bomhop, "Diagnostic Imaging Using Rare-Earth Chelates," Lasers Med. Set, 1998, H , 14-21.

80

36. M. P. Houlne, T. S. Agent, G. E. Kiefer, K. McMillan, D. J. Bomhop, "Spectroscopic Characterization and Tissue Imaging Using Site-Selective Polyazacyclic Terbium (III) Chelates;'Applied Spectroscopy, 1996, 50, (10), 1221-1228. 37. M. P. Houlne, D. S. Hubbard, G. E. Kiefer, D. J. Bomhop, "Imaging and Quantitation of a Tissue-Selective Lanthanide Chelate Using an Endoscopic Fluorometer," Journal of Biomedical Optics, 1998, 3, (2), 145-153. 38. E. Soini, T. Lovegren, "Time-Resolved Fluorescence of Lanthanide Probes and Applications in Biotechnology," CRC Crit. Rev. in Anal. Chem., 1987, 18, (2), 105154. 39. L. M. Vallarino, "Design, Synthesis and Application of Lanthanide Macrocyclic Complexes," Journal of the Less-Common Metals, 1989,149, 121-132. 40. M. Li, P. R. Selvin, "Luminescent Polyaminocarboxylate Chelate of Terbium and Europium: The Effect of Chelate Stmcture," J. Am. Chem. Soc, 1995,117, 81328138. 41. D. Parker, J. A. G. Williams, "Getting excited about lanthanide complexation chemistry," J. Chem. Soc, Dalton Trans., 1996, 3613-3628. 42. G. E. Kiefer, L. Jackson, "PRYIDALMETHYLPOLYAZAMACROCYCLOPHOSPHONIC ACIDS, COMPLEXES AND DERIVATIVES THEREOF, FOR USE AS CONTRAST AGENT," US PATENT 5,462,725; Oct. 31,1995. 43. W.D. Kim, G.E. Kiefer, J. Huskens & A.D. Sherry, "Polyazamacrocyclic fluoromonoalkylphosphonic Acids and their Complexes for Use as Contrast Agents," US application #606,162. 44. D. Parker, J. A. Williams, "Modest effectiveness of carbostyril as a sensitizing chromophore in europium and terbium amide complexes based on 1,4,7,10tetraazacyclododecane," J. Chem. Soc, Perkin Trans., 1996,2, 1581-1586. 45. A. Beeby, D. Parker, J. A. Williams, "Photochemical investigations of functionalized 1,4,7,10-tetraazacyclododecane ligands incorporating napthyl chromophores," J. Chem. Soc, Perkin Trans., 1996, 2, 1565-1579. 46. T. Gunnlaugsson, D. Parker, "Luminescent europium tetraazacyclododecane complexes with wide range pH sensitivity," Chem. Commun., 1998, 511-512.

81

47. T. Gunnlaugsson, D. Parker, J. A. Williams, "Responsive luminescent lanthanide complexes for sensing pH, p02 and hydrogencarbonate in competitive aqueous media," Part of the SPIE Conference on Advances in Fluorescent Sensing Technology IV San Jose, Califomia; January 1999: SPIE Vol. 4302: 0277786/99/$ 10.00. 48. D. Parker, N. Robert, A. Beely, " TETRA-AZA MACROCYCLES, PROCESSES FOR THEIR PREPARATION AND THEIR USE IN MAGNETIC RESONANCE IMAGING," US Patent 5,653,960; Aug. 5,1997. 49. D. Parker, J. A. G. Williams, "Getting excited about lanthanide complexation chemistry," J. Chem. Soc, Dalton Trans., 1996, 3613-3628. 50. P. R. Selvin, J. Hearst, "LUMINESCENT LANTHANIDE CHELATES AND METHODS OF USE," US Patent 5,656,433; Aug. 12,1997. 51. P. R. Selvin, J. Jancarik, M. Li, L. Hung, "Crystal Stmcture and Spectroscopic Characterization of a Luminescent Europium Chelate," Inorg. Chem., 35_, (3), 1996, 700-705. 52. M. Li, P. R. Selvin, "Luminescent Polyaminocarboxylate Chelates of Terbium and Europium: The Effect of Chelate Stmcture," J. Am. Chem. Soc, 1995,117, (31), 8132-8138. 53. P. R. Selvin, J. E. Hearst, "Luminescent energy transfer using a terbium chelate: improvements on fluorescence energy transfer," Proc. Natl. Acad. Sci. US.A, 1994, 91,(21), 10024-10028. 54. J. J. Salley, "Experimental carcinogensis in the cheek pouch of the Syrian hamster," J Dent. Res., 1954, 33, 253-262. 55. B. Heller, A. M. Kluftinger, N. L. Davis, N. F. Quenville, "A modified method of carcinogensis induction in the DMBA hamster cheek pouch model of squamous neoplasia," Am. J. Surg., 1996,172, (6), 678-680. 56. R. J. Veys, J. Baert, J. A. deBoever, "Histological changes in the hamster cheek epithelium induced by topical application of sodium lauryl sulphate," Journal Exp. Path., 1994, 75, 203-209. 57. N. Filipescu, W. F. Sager, F. A. Serafin, J Phys. Chem., 1964, 68, 3324-3346. 58. A. Doebner, W. vonMiller, Ber., 1881,14, 2816:16,1883, 2646: 1884, 17, 1699.

82

59. C. M. Leir, "An Improvement in the Doebner-Miller Synthesis of Quinaldines," J. Org Chem., 1917, 42, (5), 911-913. 60. J. A. Butera, W. Spinelli, V. Anantharaman, N. Marcopulos, R. W. Parsons, I. F. Moubarak, C. Cullinan, J. F. Bagli, "Synthesis and Selective Class III Antiarrhythmic Activity of Novel N-Heteroaralkyl-Substituted l-(Aryloxy)-2-propanolamine and Related Propylamine Derivatives," J Med Chem., 1991, 34, (11), 3212-3228. 61. R. Kubin, A. Fletcher, J. Luminesc, 1982, 27, 455. 62. J. Demas, G. Crosby, J Phys. Chem., 1971, 75, 991.

83

PERMISSION TO COPY

In presenting this thesis in partial fulfillment of the requirements for a master's degree at Texas Tech University or Texas Tech University Health Sciences Center, I agree that the Library and my major department shaU make it freely available for research purposes. Permission to copy this thesis for scholarly purposes may be granted by the Director of the Library or my major professor. It is understood that any copying or publication of this thesis for financial gain shall not be allowed without my further written permission and that any user may be liable for copyright infringement.

Agree (Permission is granted.)

Disagree (Permission is not granted.)

Student's Signature

Date