Desmosine and Isodesmosine of Elastin

Home , Desmosine, Hydroxylysine, Pyridinium

Loyola University Chicago Loyola eCommons

Master's Theses Theses and Dissertations

1973

Donald Louis Barbeau Loyola University Chicago

Follow this and additional works at: https://ecommons.luc.edu/luc_theses

Recommended Citation Barbeau, Donald Louis, "Desmosine and Isodesmosine of Elastin" (1973). Master's Theses. 2654. https://ecommons.luc.edu/luc_theses/2654

This Thesis is brought to you for free and open access by the Theses and Dissertations at Loyola eCommons. It has been accepted for inclusion in Master's Theses by an authorized administrator of Loyola eCommons. For more information, please contact [email protected].

DONALD LOUIS BARBEAU

A mESIS SUBMITTED TO TIIE FACULTY OF TIIE GRADUATE SCHOOL

OF LOYOLA UNIVERSITY OF CHICAGO IN PARTIAL FULFILLMENT OF

-,THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE

JUNE

1973

library • loyola University Medical Center ABSTRACT

11\e determination of desmosine and isodesmosine has recently been employ ed to quantitate elastin. Amino acid analysis and high voltage electrophor esis, the two methods presently used, are sensitive but time consuming when a large number of samples are to be analyzed. A preliminary investigation of the estimation of these amino acids by colorimetry and fluorimetry, sim iliar to the method for hydroxyproline in collagen, is presented. The isolation of desmosine and isodesmosine from elastin hydrolyzates was accomplished efficiently by molecular-exclusion chromatography on Bio Gel P-2 followed by ion-exchange chromatography on sulphonated polystyrene 8\ divinylbenzene cross-linked resin. Of many ion exchange resins tried, only Aminex MS "C" resin possessed the proper particle size required for the resolution of these amino acids. Nicotinamide and N(l)methyl nicotinamide were utilized as analogs of _the desmosines for the investigation of possible colorimetric techniques. The reaction of primary aranatic amines with the glutaconic aldehyde, re sulting from the destruction of the quaternary pyridinium ring, was stud ied. The reaction was sensitive to small concentrations of nicotinamide and the primary aromatic amines. The intensity of the yellow color gave a linear correlation with concentration at 415 to 425 run. The reduction of the pyridiniurn ring of N(l)rnethyl nicotinamide and isodesrnosine with sodium borohydride was performed. N(l)methyl nicotina mide exhibited a decrease in absorbance at 264 nm with a simultaneous in crease at 358 run. The absorbance of isodesmosine at 269 nm decreased. LIFE

Donald Louis Barbeau was born in Chicago, Illinois in 1948. He lived in the Chicagoland area for most of his life and attended St. Joseph High School in Westchester, Illinois. From 1967 to 1971 he attended St. Procopius College in Lisle, Illinois as a Biochemistry major. During his third and fourth years at St. Proco pius, he took part in a cooperative research program between the college's

Chemistry department and the Swift Research and Development Center in Oak brook, Illinois. As a result of the work performed in this program, a thesis and a paper were developed. The thesis fulfilled the partial requirements for the Bachelor of Science Degree at St. Procopius. The paper was presented before the 64th Annual Meeting of the Illinois Academy of Science, Collegiate Chemistry Section, Bradley University (1971); and the Associated Colleges of the Chicago Area (ACCA)'Student Symposium, Chemistry Section, North Cen tral College (1971). In the sunnner of 1971, he began his graduate research work at Hines V.A. Hospital in Hines, Illinois under the direction of Dr. A. Dietz and Dr. H. Rubinstein. In the fall of the same year, he entered the Graduate School of Loyola University of Chicago at the Stritch School of Medicine, in the Department of Biochemistry and Biophysics, located in Maywood, Ill inois. ACKNOWLEDGEMENTS

I would like to express my sincerest appreciation to both Dr. Albert Dietz and Dr. Herbert Rubinstein for supporting my work and educational expenses. I would also like to thank the members of Dr. Dietz's lab oratory for their willingness to lend a helping hand when it was needed. TABLE OF CONTENTS

. I. INTRODUCTION. PAGE A. ISOLATION AND DETECTION OF DESMOSINE AND ISODESMOSINE. 1 B. COLORIMETRIC AND FLUOROMETRIC REACTIONS OF DESMOSINE ANALOGS. 3 C. OXIDATION AND REDUCTION OF DESMOSINE AND ITS ANALOGS. 4

II. METHODS.

A. HYDROLYSIS OF ELASTIN. 6 B. PREPARATION OF BUFFERS FOR COLUMN CHROMATOGRAPHY. 6

c. PAPER CHROMATOGRAPHY OF AMINO ACIDS. 6 -· D. HIGH VOLTAGE ELECTROPHORESIS. 7

E. AMINO ACID ANALYSIS. 8

F. MONITORING EFFLUENT FROM COLUMNS. 8 - - G. ULTRAVIOLET AND VISIBLE ABSORPTION SPECTRA. 9

H. PREPARATION OF N(l)METHYL NICOTINAMIDE. 9 III. EXPERIMENTAL PROCEDURE AND RESULTS. A. FRACTIONATION OF ELASTIN HYDROLYZATES ON SEPHADEX G-15. 10 B. FRACTIONATION OF El.ASTIN HYDROLYZATES ON BIO-GEL P-2. 12 C. ION EXCHANGE CHROMATOGRAPHY. 20 D. AMINO ACID ANALYSIS. 35

-i- IV. COLORIMETRIC AND OPTICAL PROPERTIES OF 1HE DESMOSINES AND THEIR ANALOGS. 43

A. REACTION OF CYANOGEN BROMIDE WITH NICOTINAMIDE. 43

B. pH OPTIMUM FOR REACTIONS OF PRIMARY AROMATIC AMINES WITH THE GLtrrACONIC ALDEHYDE. 44 c. COMPARISON OF SENSITIVITY AMONG THE PRIMARY AROMATIC AMINES. 44

D. ABSORPTION SPECTRA OF POLYMETHINE DYE FORMATION. 51

E. SODIUM BOROHYDRIDE REDUCTION OF N(l)METHYL NICOTrnAHIDE. 51 F. SODIUM BOROHYDRIDE REDUCTION OF ISODESMOSINE. 56

G. REDUCTION OF N(!)METHYL NICOTINAMIDE WITH SODIUM DITHIONITE. 56 v. DISCUSSION.

-ii- LIST OF FIGURES

FIGURE DESCRIPTION PAGE 1 STRUCTURE OF DESMOSINE AND ISODESMOSINE 2 2 FRACTIONATION OF HYDROLYZATES ON SEPHADEX G-15 11

3 FRACTIONATION OF HYDROLYZATES ON SEPHADEX G-15 13 4 FRACTIONATION OF HYDROLYZATES ON BIO-GEL P-2 16

5 AMINO ACID ANALYSIS OF FRACTION IN ON PA-35 17

6 FRACTIONATION OF HYDROLYZATES ON BIO-GEL P-2 18

7 AMINO ACID ANALYSIS OF FRACTION Io ON PA-35 19

8 FRACTIONATION OF HYDROLYZATES ON Q-15S 22

9 ELUTION OF IN FROM AMINEX MS "C" WITH 0. 20 ! CITRATE pH 4.30 24 -10 ELtn'ION OF Io FROM AMINEX MS "C" WITH 0.20 li CITRATE pH 5.28 26

11 ULTRAVIOLET ABSORPTION SPECTRA OF ISODESMOSINE FROM AMINEX COLUMN 27 12 ELUTION OF Io FROM AMINEX MS "C" WITH 0. 20 li CITRATE pH 4.30 at 56°C 28

13 AMINO ACID ANALYSIS OF DESMOSINE FRACTION (4c) ON PA-35 29 14 AMINO ACID ANALYSIS OF ISODESMOSINE FRACTION (4b) ON PA-35 30 15 ULTRAVIOLET ABSORPTION SPECTRA OF DESMOSINE (4c) AND ISODESMOSINE (4b) 31 16 FRACTIONATION OF HYDROLYZATES ON AMINEX MS "C" WITH 0.20 li CITRATE pH 4.30 AT 56°C 32

-iii- FIGURE DESCRIPTION PAGE 17 ELUTION OF le FRACTIONS FROM AMINEX MS "C" WITii 0.38 !!_ ITRATE pH 4.30 34 18 AMINO ACID ANALYSIS OF EI.ASTIN HYDROLYZATE ON PA-35 36 19 AMINO ACID ANALYSIS OF BASIC AMINO ACID MIXTURE ON PA-35 37 20 AMINO ACID ANALYSIS OF BASIC AMINO ACID MIXT~RE ON PA-35 38 21 AMINO ACID ANALYSIS OF ELASTIN HYDROLYZATE ONAA-15 39 22 PAPER CHROMATOGRAPHY AND HIGH VOLTAGE ELECTRO- PHORESIS OF ELASTIN HYDROLYZATES 42 23 REACTION OF CNBR WITH NICOTINAMIDE 46 24 FORMATION OF POLYMETHINE DYE WITH DESMOSINE AND PRIMARY ARrn.fATIC AMINE 47 25 REACTION OF CNBR WITH NICOTINAMIDE AT THREE TEMPERATURES 48 26 COMPARISON OF SENSITIVITY AMONG VARIOUS PRIMARY AROMATIC AMINE WITH GLUTACONIC ALDEHYDE so 27 ABSORPTION SPECTRA OF £,-PHENYLENEDIAMINE AND CYANOGEN BROMIDE COMPLEX 53

28 ABSORPTION SPECTRA OF E,1J:?.-AMINODIPHENYL AND CYANOGEN BROMIDE COMPLEX 54 29 ABSORPTION SPECTRA OF REDUCED AND NON-REDUCED N(!)METHYL NICOTINAMIDE IODIDE 55 30 ABSORPTION SPECTRA OF REDUCED AND NON-REDUCED ISODESMOSINE 57

-iv- LIST OF TABLES

TABLE DESCRIPTION PAGE 1 AMINO ACID COMPOSITION OF SEPHADEX G-15 FRACTIONS 14

2 AMINO ACID COMPOSITION OF BIO-GEL P-2 FRACTIONS 21

3 AMINO ACID ANALYSIS OF ELASTIN HYDROLYZATE ON AA-15 AND PA-35 RESINS 40 4 MOBILITY OF A\f INO ACIDS ON PAPER CHROMATOG- RAPHY AND HIGH VOLTAGE ELECTROPHORESIS 41 5 Sl'RUCTURE OF PRIMARY AROMATIC AMINES 45

6 OPTIMUM pH FOR AROMATIC AMINES FORMING POLYMETHINE DYES 49

7 REDUCTION OF N(!)METHYL NICOTINAMIDE WITH 52 ------SODIUM BOROHYDRIDE

-v- INTRODUCTION

During an investigation of the components responsible for the cross linking in elastin, two novel amino acids of large molecular weight were discovered by Partridge (1). The two compounds, later designated desmo sine and isodesmosine from the Greek 'desmos' meaning to join, are tetra aminotetracarboxylic acid isomers containing a pyridinilDll ring as their

nucleus (2,3).

As the structures indicate in figure 1, the four «-amino carboxylic acid residues present are responsible for the four-fold enhancement of the ninhydrin color yield. This facilitates their detection as amino acids, while their ultraviolet absorption fesulting from the quaternary ring al lows them to be characterized more specifically. A complete review of the structure, properties, and early work on the biosynthesis of the desmosines has been compiled (3-5). The isolation of desmosine and isodesmosine from elastin hydrolyzates has mainly been accomplished by elution from sulphonated polystyrene, di vinylbenzene crosslinked ion exchange resins (1,6,7). Initially, the desmosines were eluted together from either allDllina or the sulphonated resins. Large volwnes of 1.5 to 4.0 ! HCL are required for the latter technique. Employing the same resin and various concentrations of sodilVll citrate buffers, the desmosines were resolved as discrete ninhydrin and ultraviolet absorbing peaks (1,3). 1be resolution of these isomeric amino acids with citrate buffers has made possible their quantitative detennination and isolation on amino acid analyzers (8). Due to their atypical behavior on these resins,

- 1- Desmosine lsodesmosine

Figure 1 modifications in buffer and colunm height had to be made (9-17). The two-collDlln procedures employed are similiar to physiological nms (9-11). One-colwnn procedures have proven effective (12-16), as well as gradient elution techniques on single columns (17). In order to facilitate the isolation of large quantities of the des mosines or their detection in tissue, molecular exclusion on Sephadex (7) and gel filtration on Bio-Gel (18,19) have been successfully employed. Although the desmosines were eluted together with these methods, as des cribed for the hydrochloric acid elution on ion exchange resins, the speed and ease with which this can be accomplished are improved. Subsequent e lution from ion exchange resins with sodilDll citrate buffers has provided a valuable technique for not only isolating the isomers, but also calcu lating their content in tissue. High voltage electrophoresis has shown to be adequate for the deter mination and isolation of the desmosines (20). Used as an analytical tech- .. nique, the ninhydrin treated electropherogram can be used to quantitate the desmosines by their absorbance in a densitometer. By cutting out the areas

containing the desmosines and eluting on a colwnn with o.so ~ HCl, it is possible to obtain the isomers on a preparative scale, unresolved. In the same way that hydroxyproline detennination is used as a diag nostic test for collagen diseases (21-23), it has been proposed that the desmosines be used to quantitate elastin. The only methods available to date are high voltage electrophoresis and automated amino acid analysis. A colorimetric method, similiar to the one used for the detection of hydroxyproline, would be far more convenient. Geyer (24) has suggested that the desmosines are responsible for the staining of elastic tissue when treated with Morel-Sisley reagents. The Morel-Sisley reaction has been adopted as a diazonium-coupling reagent for the detection of protein-bound tyrosine (25-27). Adaptation of this reac tion to the specific detection of the desmosines would be of considerable value, however its specificity has not been established.

A more direct approach to the detection of the desmosines would take advantage of their unique pyridinium ring. Several colorimetric and fluorometric reactions for nicotinic acid derivatives, analogs containing this ring, have already been established (28-30). These involve the for mation of a glutaconic aldehyde from the destruction of the pyridinium ring with sodium hydroxide, with the subsequent fonnation of a polymet hine dye in the presence of a primary aromatic amine (31). The possi bility of adapting such a reaction to the quantitation of the desmosines is supported by evidence that these amino acids behave similarly to the N(l)methyl nicotinic acid derivatives when the ring is destroyed with sodium hydroxide (32).

Oxidation of N(l)methylnicotinamide with K3Fe(CN) 6 produces highly fluorescent 2 and 6 pyridones (33-36). The absorption spectra of these carbonyl adducts can be used as a diagnostic tool. Similar results should be obtained with desmosine, giving a method for its quantitation.

1be reduction of the pyridinium ring in the desmosines with NaBH4 has been previously established (37-41). This affords a change in the optical properties of the two compounds which then resemble a class of

-4- compounds known as the dihydropyridines. A complete review of the phys ical and chemical properties of the dihydropyridines is presently avail able (42). Several treatises on the mechanisms of NaBH4 and sodium dithionite reduction of pyridinillll compounds are also available (43-45). The present paper presents methods of isolation and characterization of desmosine and isodesmosine, along with a preliminary investigation concerning their quantitative detennination by the methods outlined a bove.

-5- METOODS

HYDROLYSIS OF ELASTIN In a sealed tube, purified elastin obtained from Worthington Biochem

ical Company was hydrolyzed in 6 ~HCl (SOmg elastin/10 ml) for 72 hours at 110°C under nitrogen. The hydrochloric acid was removed by repeated evaporation on a rotary evaporator until the pH of the solution was 1.5 to 2.0. The resulting solution was evaporated to dryness and the residue dissolved in the appropriate buffer. When preparing hydrolyzates for the Bio-Gel P-2 colunn, the pH was adjusted to 2.8 with dilute NaOH , treated with Norit, and filtered prior to its application (18).

PREPARATION OF BUFFERS FOR COLUMN CHROMATOGRAPHY o.os Molar pyridine-formate buffer pH 3.00. To 1800 ml of distilled water, 8.0 ml (7.91 g) pyridine and 18.3 ml (21.92 g) of 88% formic acid are add ed to give a pH of 3.oo. Distilled water is added to 2 liters. - Sodium citrate buffers. The procedure outlined in the Beckman Amino Acid Analyzer Manual was followed for the preparation of these buffers. The only exceptions were the elimination of thiodiglycol and the substitution of chloroform for caprylic acid as preservative. (NOTE: Normality refers to the sodium concentration)

PAPER CHROMATOGRAPHY OF AMINO ACIDS All paper chromatography was performed in a cylindrical glass contain er (14 x 44 cm) on Whatman #1 filter paper in an ascending manner. After development in the appropriate solvent, the chromatogram was allowed to dry

-6- thoroughly before dipping in a 0.2% ninhydrin, 1% acetic acid-acetone solution. Again the chromatogram was allowed to dry and placed in an oven at 110°C for S to 10 minutes to hasten the ninhydrin color reaction. n-Butanol/acetic acid/water 4:1:5. A solution of 250 ml of these solvents was prepared in a separatory funnel and shaken vigorously for 30 to 60 sec onds. After allowing to layers to separate completely, the lower aqueous layer was placed in a small beaker on the bottom of the cylindrical jar. Equilibration proceeded for 30 minutes prior to and during development. Approximately 50 ml of the liquid from the upper layer was used as the developing solvent and was placed directly on the bottom of the jar. After 16 hours, the paper was removed and treated as outlined above (7). Pyridine/acetic acid/water 30:20:50. A 50 ml volume consisting of these solvents placed on the bottom of the chromatography jar was sufficient for

a 16 hour T\D'l. 1he resulting chromatogram was treated as above. n-Butanol/acetic acid/water 35:30:35. Sixty ml of this solvent was suffic- -- ient to develop the chromatogram. 1he length of the run was 42 hours after which time the chromatogram was treated as above (19).

HIGH VOLTAGE ELECfROPHORESIS All electrophoresis was carried out in a Camag HVE cell on Schleicher &Schuell 20408 20 x 40 cm paper. Application of samples 6 inches from the left edge (positive pole) was sufficient for the length of the run. After completion, each electropherogram was dried in an oven at 110°C for S min utes. 1he paper was sprayed with ninhydrin and allowed to air dry, followed by 2 to 3 minutes in the oven at 100°C to hasten color development.

-7- pyridine/acetic acid/water 1:10:89 Parameters for this buffer were 3000 volts, l~O mA, for 26 minutes (20). Formic acid/acetic acid/water 5:12:383 pH 2.1 This system was used at 4300 volts, 75 mA, for 15 minutes. 8% formic acid pH 1.6 For this system, 4000 volts, 160 mA, for 15 min utes was sufficient (46).

AMINO ACID ANALYSIS A modified Beckman 120 C model amino acid analyzer was used with methods for the basic and total amino acids, and elastin hydrolyzates. Basic column PA-35 resin at a height of 15 cm, and flow rates of 68 ml/hr for buffer and 35 ml/hr for ninhydrin proved sufficient for the separation of the basic components of the elastin hydrolyzates. A 0.20 N sodiwn ~ citrate buffer pH 4.25 was employed for the first 40 minutes followed by a 0.35 !!_buffer pH 5.28 thereafter, all at 53.5°C. Single column-total hydrolyzates AA-15 resin at a height of 60 cm, and flow rates of 70 ml/hr for buffer and 35 ml/hr for ninhydrin were adequate for complete separation of the amino acids of ~elastin hydrolyzates with the exception of the desmosines. The desmosines resolved, but were ob- scured by the tyrosine peak. A 0.20 N sodium citrate buffer pH 3.49 was used for the first 40 minutes followed by 0.40 !!_buffer pH 4.12 which was again changed at 70 minutes to a 1.0 !!_buffer pH 6.40. The temperature throughout the run was maintained at 53.5°C (47).

MONITORING EFFLUENT FROM COLUMNS Individual fractions collected either by a siphon device (volume)

-~ or photocell (dropwise) were monitored for ultraviolet absorption with a Beckman DU Spectrophotometer equipped with matched quartz cells. Their ninhydrin reaction, utilizing the method of Rubinstein (48), was measured with a Coleman Model 6-20A Spectrophotometer.

ULTRAVIOLET AND VISIBLE ABSORPTION SPECTRA

A Beckman DB Spectrophotometer with a Seargent SR Recorder was used.

PREPARATION OF N(l)METHYL NICOTINAMIDE IODIDE

To 30 ml methanol was added 4.0 g nicotinarnide, 3.5 ml methyl iodide, and a boiling chip. Tilis mixture was refluxed S hours at 60°C, cooled in an ice bath, filtered, and the residue washed with cold methanol. An 83% yield

(7.2 g) was obtained and the ultraviolet absorption spectra compared to a sample of N(l)methyl nicotinamide iodide purchased from Sigma Chemical Com pany (33).

-9- EXPERIMENTAL PROCEDURE AND RESULTS

FRACTIONATIO~ OF ELASTIN HYDROLYZATES ON SEPHADEX G-15

Elastin hydrolyzate (100 mg) was dissolved in 1.7 ml of 0.05 !:!, pyridine-fonnate buffer pH 3.00 and applied to a Sephadex G-15 column. 1be 2.6 x 61 cm colunn, which was rinsed with this buffer prior to use, had a flow rate of 150 ml/hr via gravity flow. Using a volumetric siphon device and fraction collecter, 5 ml fractions were collected and monitor ed by their absorption a~~ 274 nm. The elution profile in figure 2 indi cates several regions exhibiting u.v. absorption, although the fastest fraction should contain the high molecular weight desmosines. Blue dex tran 2000 was used as a void volume marker. 1be fastest fraction eluted, IJ, was analyzed by paper chromatography and high voltage electrophoresis and shown to contain the desmosines. Further division of this fraction was carried out and analyzed by amino acid analysis on PA-35. The results of this analysis are represented by fraction 11, to be presented later. Analysis of the second u.v. absorbing peak by amino acid analysis using the single column technique (AA-15) revealed that the major portion of the amino acids are found in fraction IIJ. Although proline is found in this fraction, it is present in only a very small amount ccmpared to the remaining amino acids. The bulk of the pro line can be found as part of the ( IJ ) fraction. High voltage electrophoresis has verified the presence of the more abundant amino acids in the IIJ fraction. The third major u.v. absorbing region, fraction IllJ' was subjected

-10- 2.00

Q) 1.00 ;u

-e0 VI ~

I ..... •

140 180 220 260 300 340 ml effluent Figure 2 Elution of elastin hydrolyzates from Sephadex G-15. Fraction IJ (115-155 ml), fraction IIJ (205-275 ml), fraction IIIJ (280-315 ml). Void voltDne=llO ml. to amino acid analysis and found to contain only the aromatic constituents. Tyrosine and phenylalanine are known to be retarded on Sephadex gels due to the affinity of their aromatic1>electron system to the gel matrix (18, 49-51).

On the same column, 200 mg elastin hydrolyzates (1 ml) were eluted as prescribed above in order to analyze the fastest fraction further. 11le flow rate of the column was 72 ml/hr. Again S ml fractions were collected as before and monitored at 274 nm. 1he elution pattern in figure 3 repre sents the fastest fraction collected. Further division of the peak re vealed a better separation of the desmosine components. Peak II contain ed the desmosines with a trace of ornithine andl'-aminobutyric acid when analyzed with the amino acid analyzer (PA-35).

Fraction II1 contained large amounts of valine, alanine, glycine, and proline. Although table 1 indicates the presence of many other amino acids, their concentration is practically negligible compared to these four. Both high voltage electrophoresis and amino acid analysis were used to identify the constituents of this fraction.

FRACI'IONATION OF ELASTIN HYDROLYZATES ON BIO-GEL P-2 Elastin hydrolyzates (700-800 mg) were treated as prescribed in met hods and 3 ml applied to a Bio-Gel P-2 column. Before use, the l.S x 88

cm column was first rinsed with o.s ~acetic acid and a 0.20 ml mixture of markers eluted. Cytochra11e C was used as void volume marker and cyano cobalamin to mark the position of lysine (19). Blue dextran was not used due to its affinity to the gel matrix (20). 1he hydrolyzates were then

-12- 2.00 ••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••

G> au -e g 1.00 ~

.....I ~ I

120 140 160 180 200 220 ml effluent

Figure 3 Elution of elastin hydrolyzates from ~ephadex G-15. Fraction 11 (135- 155 ml ), fraction 11 1 (160-220 ml ). Dotted line indicates nifihydrin reactive fractions. Void voltnne=llO ml. TABLE 1

AMINO ACID COMPOSITION OF SEPHADEX G-15 FRACf IONS

AMINO ACID IJ II IIJ III IIIJ

Aspartic acid B B Threonine B B Serine B B Glutamic acid BA B Pro line F BA BA Glycine BA BA Alanine BA BA Valine BA BA Isoleucine B B Leu cine B B Arginine A A Desmosine CD EF Isodesmosine CD EF Tyrosine B Phenylalanine B Methionine A Hydroxyproline B Ornithine E B Hydroxy lysine B Lysine BA Histidine It -aminobutyric acid E

Letters represent method of analysis. (A) High voltage electrophoresis with 8% fonnic acid at 4000 volts, 170mA, for 15 minutes. (B) Amino acid analysis employing the AA-15 resin method. (C) Paper chromatography with pyridine/acetic acid/water 30:20:50. (D) High voltage electrophoresis with pyridine/acetic acid/water 1:10:89 pH 3.74 at 4000 volts, 120 mA, for 26 miriutes. (E) Amino acid analysis employing PA-35 resin method. (F) Paper chromatography in ,!!.-butanol/acetic acid/water 4:1:5

-14- applied to the column and eluted with 0.5 ~acetic acid by gravity flow. I Using a photocell, 2 ml fractions were collected dropwise and monitored for their absorption at 274 run.

Only two u.v. absorbing regions were eluted as shown in f~gure 4. The first region, IN, was analyzed by paper chromatography, amino acid analysis, and high voltage electrophoresis. As shown in figure 5, from the amino acid analysis on PA-35, this region not only contains the desmosines, but other basic amino acids as well. Glycine and alanine were also found in this region. The second region, IIN, contained the acidic and neutral amino acids. Proline was also eluted in this region. 1he amino acid composition of the area between the two peaks consisted of lysine and other basic amino acids, as well as a small amount of valine. A larger colwnn was prepared in an attempt to improve the separation of the desmosines from the remaining amino acids, and the above procedure repeated. Elastin hydrolyzate (l.O g) was applied to a 2.6 x 89 cm colwnn of Bio-Gel P-2 after calibration of the column with cytochrome C and cyanocobalamin (5 ml). Prior to elution, the column was rinsed with the eluant (0.5 ~acetic acid). 1he 4 ml hydrolyzate preparation was eluted at a flow rate of 171 ml/hr, and the fractions collected using a 5 ml siphon device. As shown in figure 6, the elution of two u.v. absorbing regions was

accomplished. The first and the most intense region, I0, was comprised mainly of desmosine, isodesmosine, and ornithine as depicted by the amino acid analysis on PA-35 shown in figure 7. Paper chromatography in n-butanol/acetic acid/water (35:30:35) allowed the desmosines to move

-15- 1.00

II N

ll! 'o:t "N I I Cl) 0.20 u I s:: 0.50 c'3 I

-e0 I' ! Ill I \ 0 ~ 00 I \ N .....I 0.10 CJ'. I \ I I \ \ \ IIIN \ ' ,_ 20 40 60 80 100 ml effluent Figure 4 Elution of elastin hydrolyzates from Bio-Gel P-2. Fraction IN {50-63 ml), fraction IIN {73-87 ml ), fraction IllN {65- 71 ml). Void volume=44 ml with 0.40 J..uDOles cytochrome C and lysine marker=57 ml with 0.16 µmoles cyanocobalamin. c... 0

in u "O 0 in ...a -..::J ..in z 0 c:O u "O u _A•0.3 <(

m. I . ; .....,...... ;; I :1. >.·' . . . ::c:. ' .. : ~ in !t· >. rt> : ~ : ::. ·" ., ...J :::c: • • : \ ,J... ~ ! \' .· h z ~:__'_,,,,,_!)- ______~\"-\\,...>,--=::"======_)r j \~

10 30 l:>c 50 70 90 min 110 130 150 170

Amino acid analysis of fraction I on PA-35 resin. GAB=~-aminobutyric acid. Figure 5 N 2.00

0.30 I I e I ...... ~ I N I '\ ~ Q) 1.00 I 0.20 u \ 0 s:: ( 00 (IS \ N -e r \ 0 Q) Cl) f I u ~ I a ' II0 (I IQ) -e I 0.10 0 ...... ' Cl) 00 I' I ~ I \ I \ I , ' ' '

200 250 300 350 400 450 ml effluent

Figure 6 Elution of elastin hydrolyzates from Bio-Gel P-2. Fraction 10 (195- 260 ml), fraction 110 (290-370 ml). Void voltune=l90 ml with 0.08 µmoles cytochrome C and lysine marker•251 ml with 0.32 µmoles cyano cobalamin...... "°0 -...... ,o _A•0.3 .. ::

...c I 0 ...... t ,. l.O m. I 1; ~ • >. : : r.J')

=:::::i--~...:======-;'Jv~_:c_,,;:::;;::_z~------r - 10 30 be 50 70 90 min 110 130 150 170

Figure 7 Amino acid analysis of fraction 10 on PA-35 resin. •

as a discrete spot with a mobility o.46 that of histidine. Region II was analyzed by collecting a corresponding fraction from 0 a subsequent run (Ilq) and subjecting it to amino acid analysis. It consisted mainly of the acidic and neutral amino acids with tyrosine and phenylalanine again being retarded as they were on the Sephadex column. Table 2 illustrates the methods of analysis used to identify the canponents of each of these fractions.

ION EXCHANGE OlROMATOGRAPHY A 0.9 x 13 cm column of Q-155 (particles 19-25 µ in diameter) ion exchange resin was first regenerated with 0.20 !NaOH and equilibrated with 0.20 !i sodium citrate buffer pH 5.28 before use. A O.S ml calibra - tion mixture of tyrosine, lysine, and phenylalanine were then eluted in the manner described below. Subsequently, elastin hydrolyzates (2 mg/0.25 ml) were applied and the amino acids eluted with 0.20 !i sodium citrate buffer, pH S.28. lbe flow rate of the buffer through the column was 66 ml/hr with a Technicon peristaltic pump while the temperature was main tained at S0°C. For the calibration mixture, 1 ml fractions were collected dropwise and two ml fractions for the hydrolyzates. Each was monitored with ninhydrin as specified in the section dealing with methods.

Figure 8 shows the elution profile of both the hydrolyzates and the calibration mixture. Peaks la and lb represent the acidic and neutral amino acids, and lysine and phenylalanine respectively. 'Ibis is an in- sufficient separation of the basic amino acids since the lysine elution volume is the same as phenylalanine. Paper chromatography using !!,.-butanol/ acetic acid/water (4:1:5) was employed to analyze the components of each -20- TABLE 2 AMINO ACID COMPOSITION OF BIO-GEL P-2 FRACTIONS

AMINO ACID IN Io IIIN IIN IIq

Aspartic acid B B Hydroxypro line B B Threonine B B Serine B B Glutamic acid B B Pro line B B Glycine AF B B Alanine AF B B Valine B B B Isoleucine B B Leu cine B B Methionine B Desmosine AEF AEF Isodesmosine AEF AEF Ornithine AE AE B a"-aminobutyric acid E E Histidine A AE Hydroxy lysine E E B Lysine EF B

Letters represent method of analysis. (A) High voltage electrophoresis with 8% formic acid at 4000 volts, 150 mA, for 15 minutes. (B) Amino acid analysis employing AA-15 resin method. (E) Amino acid analysis employing PA-35 resin method. ' (F) Paper chranatography with ~butanol/acetic acid/water 4:1:5.

-21- 2.00 la

e 0 "'U') Cl> 1.00 ,, s::0 ell -e ) ' 0 I l lb UI I I ~ I I N N I ' J I -,,, J

20 40 60 80 100 ml effluent Figure 8 Elution of elastin hydrolyzates from Q-lSS ion exchange resin. Calibration mixture superimposed consists of 0.3 µmoles tyrosine, 0.1 µmoles phenylalanine, arnd O.l µmoles lysine in 0.20 N soditun citrate buffer pH 3.25. Dotted line represents-calibration mixture. peak. Similiar results were obtained for 4 mg of hydrolyzates and the calibration mixture when 0.20 N sodium citrate buffer pH 4.30 was used as eluant. After regenerating an Aminex MS "C" (particles 33-47 µ in diameter) resin with 0.20 !!_NaOH and equilibrating with 0.20 !!_sodium citrate buf fer pH 4.30, fraction IN was applied to the 1.2 x 47 on column. Tile new column had a flow rate of 27 ml/hr with a Technicon pump and was operated at room temperature. Tile sample was eluted with the 0.20 ~citrate buffer and fractions collected dropwise. For the IN sample, 2.0 ml fractions were collected, while 2.3 ml fractions were collected for the calibration mixture. The elution was tenninated after 360 ml had been collected with only two ninhydrin peaks appearing. Peaks 2a and 2b were analyzed by paper chromatography with Jl:butanol/acetic acid/water (4:1:5) and found to be glycine and alanine. The calibration mixture was eluted as one peak half way through the run. The two runs are superimposed in figure 9. Desmosine and isodesmosine were not eluted with this volume of effluent.

On the same resin as above, 0.5 ml of fraction I0 was eluted with 0.20 !!_sodium citrate buffer pH 5.28. The column, which had been altered to 1.5 x 82 on of resin, was operated at room temperature. Before elution of the sample, the resin was washed with 0.20 N NaOH and the eluting buf fer as previously described. Tile flow rate with the Technicon pump was 26 ml/hr. Tile 2.4 ml fractions collected dropwise were monitored both by their absorption at 274 nm and with ninhydrin. Tile column was allowed to operate until 480 ml effluent was collect-

-23- 4.00 2a

l! 0 r-. LI) 2.00 4) 0.20 CJ ,, fj -e I \ 0 Ill ' \ ~ 2b r ' I 0 .10 N .j::o • l I ' ' I \ I \ J \ I - '-. .... ------..- 60 120 180 240 300 360 ml effluent Figure 9 Elution of IN from Aminex MS "C" with 0.20 N sodiun citrate buf fer pH 4.30. Calibration mixture (----) contains 0.3 JllllOles ty rosine, 0.1 µmoles phenylalanine, and 0.1 vmoles lysine in 0.20 N citrate buffer pH 3.25. ed. The elution profile in figure 10 indicates two ninhydrin peaks with one possessing considerable U.V. absorption. The two peaks were analyzed by paper' chromatography with n-butanol/acetic acid/water (4:1:5) to ident ify 3a as either glycine or alanine, and 3b as desmosine or isodesmosine. Further identification of 3b showed that: high voltage electrophoresis with fonnic acid/acetic acid/water (5:12:383) gave a spot with a mobility of 0.61 relative to lysine; amino acid analysis with PA-35 indicated iso desmosine; and the ultraviolet absorption spectra of the non-desalted fraction corresponded to that of isodesmosine (figure 11). Alterations of coltnnn size were again made along with provisions to operate the column at 56°C. The Aminex MS "C" resin was regenerated with NaOH and the eluting buffer as described earlier. The o.9 x 26 cm coltDDn of resin had a flow rate of 17 ml/hr with the Technicon pump. A 0.5 ml aliquot of fraction 10 in 0.20 ! sodium citrate buffer pH 3.25 was applied to the column and eluted with 0.20 ! sodium citrate pH 4.30. The 1.4 ml fractions, collected dropwise, were monitored by both their ninhydrin reaction and their absorl>ance at 274 nm. Figure 12 clearly indicates two ninhydrin peaks of equal intensity corresponding to two U.V. absorption peaks. Analysis of 4b and 4c by high voltage electrophoresis with 8% formic acid showed both desmosine(4c) and isodesmosine (4b) as spots with a mobility 0.75 that of lysine. Amino acid analysis with the PA-35 resin method produced not only an identifica tion of each peak, but also a check on their purity. Figures 13 and 14 show that although the desmosines are the major constituents of the two synunetrical peaks fran the Aminex resin, there is a small amount of con tamination by other basic amino acids. This contamination is almost neg-

-25- 2.00

e 0 "LJ')

Q) u 1.00

a I of I 0 0.40 I Ill 3b I ~ ~ "N

I Q) N 0\ u I 3a 0.20 a ~ 0 Ill ~

80 160 240 320 400 480

ml effluent

Figure 10 Elution of I0 fran Aminex MS "C" with 0.2 N sodium citrate buf fer pH S.28. 0.40

o.3o

Q) () c:: as

-e0 0.20 Ill ~

0.10

240 280 320 nm

Figure 11 Ultraviolet absorption spectra of peak 3b from Aminex column.

-27- 2.00 4b 4c

~ ...... 0 U')

Q) u c:: 1.00 1.00 ell i: 0 Ill ~ 4d ~ ...... ~ 4a N I o.so N Q) 00 u I c:: c1S i: 0 Ill ~

14 28 42 56 70 105 112 ml effluent

Figure 12 Elution of 10 from Arninex MS "C" with O. 20 N sodium citrate buffer pH 4.30 at 56°C. Peaks 4b and 4c are isodesmosine and desmosine respectively. A=0.3

(/) cu I Cl N \0 I t: : ..: :

----~A~~~~NCU~-

60 80 min 100 12.0

Figure 13 Amino acid analysis of fraction 4c on PA-35 resin. Peak at 95 minutes is unidentified. A=0.3

I t#.1 0 I

60 80 min 100 120

Figure 14 Amino acid analysis of fraction 4b on PA-35 resin. Peak at 95 minutes is unidentified. 0.40 . 0.40

o.3o 0.30

0.20 0.20

i 0.10 ...... 0.10 I

240 280 320 tun 240 280 320 tun ISODESMOSINE (4b) DESMOSINE (4c)

Figure 15 Ultraviolet absorption spectra of desmosine and isodesmosine collected from Aminex MS "C" resin. 2.00 Sa

e r--0 LI)

Q) Sb u s:: 1.00 0.10 -ellS 0 Ill ~ o.os I Sc Sd (.t.I N I I -\ I \

14 28 42 S6 70 84 ml effluent Figure 16 Elution of elastin hydrolyzates from Aminex MS "C" with 0.20 N sodium citrate buffer pH 4.30 at S6°C. Peaks Sc and Sd are the desmosines. ligible. The ultraviolet absorption spectra of the two compounds shown in figure IS.correspond to those reported in the literature (3). Frac- tion 4d was analyzed by amino acid analysis on PA-35 and showed a single peak corresponding to ornithine. 11lis would indicate a satisfactory separation of the basic amino acids of elastin on this resin under the above operating conditions. Under identical conditions, 4 mg elastin hydrolyzate, dissolved in o.s ml 0.20 !!, citrate buffer pH 3.25 was fractionated. 11le elution profile shown in figure 16 illustrates the position of the desmosines with respect to the acidic, neutral and basic amino acids, (Sa and Sb respectively). Isodesmosine (Sc) and desmosine(Sd) were eluted in the same position as before.

ln~an attempt to isolate desmosine and isodesmosine on a larger scale, a 2.6 x 47 cm column of Aminex MS "C" was prepared. Before use, the resin was treated with NaOH and eluant as described earlier. A 6 ml ( 6 mg hydrolyzate) sample of combined fractions corresponding to 10 from the Bio Gel column was eluted with 0.38 !!. sodium citrate buffer pH 4.30. 11le col- , umn temperature was maintained at S6°C •. Using a S ml siphon, fractions were collected and monitored by their absorption at 274 nm. 11le elution profile in figure 17 indicates several regions absorbing in the u.v •• 11le isodesmosine fraction (6b) was analyzed by amino acid analysis on PA-3S and revealed several other amino acids, including ornithine. Fraction 6c contained 2S"-aminobutyric acid, desmosine, isodesmosine, hydroxylysine, and ornithine as shown by amino acid analysis on PA-3S. -33- 6b 2.00

«> g 1.00 6c "' -e0 Ill ~

I ti-I ~ I

50 100 150 200 250 ml effluent Figure 17 Elution of fastest fractions from Bio-Gel column from large Aminex MS "C" column with 0.38 N sodium citrate buffer pH 4.30. AMINO ACID ANALYSIS 1he analysis of the basic amino acids and the desmosines was perform ed on the 15 cm colwnn of PA-35 resin. Figure 18 illustrates the resolu tion of these components when an elastin hydrolyzate was applied to the column. Prior calibration of the column was achieved with two mixtures of basic amino acids as shown in figures 19 and 20. 1he analysis of the complete elastin hydrolyzate was accomplished using the single-colwnn method of Kremen and Vaughn (47). As shown by figure 21, this method was inadequate for the resolution of the desmosines. This column was calibrated with a Beckman amino acid standard.

-35- .-. .. ' ! ·:: : . D .:-:·: ~ .2 ...... -1. ~ -0 . . a.. ~ co... > J- '

,,, -. A•0.3 xz ..... -0 ... 0 .. I !?o (I.I :· ·.•. ·· .t °'I .. ~ :: ., ' I

. : •'

-=iI 10 30 be 50 70 90 min 110 130 150 170

Figure 18 Amino acid analysis of elastin hydrolyzate (O.S mg) on PA-35 resin. _A•0.3

~ .!! ~ 0 .. ::c !: ...J""' ·' t~ !~ A. : ...... : : :'\ ! 1 : :

I ...... ,~ I ;:;:;;;;; ;::z;;c;=Gi;/v.~-~=====~d--~\JJ)p. J 10 30 be 50 70 90 min ltO 130 150

Figure 19 Basic amino acid mixture (0.05 µmole each) on PA-35 resin. A•0.3

a. >. ::c

.. I (,,.I 00 .1''t·· I ======:::::::=::::::;::.,1 ~~.:::::::;:::=~'./\--:\:::::==:::;:::====::;::::::=:::==;::::=

10 30 be· 50 70 90 min 110 130 150 170

Figure 20 Basic amino acid mixture (0.05 µmole each) on PA-35 resin...... ! ..!!°. D -:' .. er. > · . ..J..

.. ,.., ::! . :r .. z .&. .o.

-. A•0.3 .a 0 (!) Ir. ' .

Q. :l :: t .. 0 (/) ..

I. ~ Q. :: :· :::l' ::- .. -=:::::====::1.1li:Ju1u 20 40bc 60 be 80 100 min 120 140 160 180 200

Figure 21 Amino acid analysis of elastin hydrolyzate (O.S mg) on AA-15 resin. TABLE 3

AMINO ACID ANALYSIS OF ELASTIN HYDROLYZATE ON AA-15 AND PA-35 RESINS

AMINO ACID M-15 RETENTION TIME PA-35 (MIN)

Aspartic acid 39 Hydroxyproline 41 IS Threonine 48 Serine 51 Glutamic acid SS Prolin e 63 Glycine 77 Alanine 81 Valine SS Methionine 93 Isoleucine 9S Leu cine 100 1'-aminobutyric acid 111 71 lsodesmosine 114 74 Desmosine 116 77 Tyrosine 116 29 Phenylalanine 121 31 Omithine 129 91 Histidine 139 105 Hydroxy lysine 146 84 Lysine 149 97 Ammonia 177 107 Tryptophane 77 Arginine 166

-40- TABLE 4

MOBILITY OF AMINO ACIDS ON PAPER CHROMATOGRAPHY AND HIGH VOLTAGE ELECfROPHORESIS

Ra Rb b c AMINO ACID f f Rieu Rtys

Glutamic acid .Sl Alanine .28 .65 Valine .56 .s2 Glycine .23 • 71 Serine .24 Threonine .27 Pro line .33 .55 Isoleucine .63 Leucine .66 .so, 1.00 Hydroxyproline Phenylalanine .59 .so 1.00 Tyrosine .42 • 74 .93 Histidine .12 .60 • 75 .82 Arginine .15 .89 ~~aminobutyric acid .95 Lysine .10 .61 • 76 1.00 Hydroxy lysine 1.02 Ornithine .97 Desmosine .01 .46 .58 • 75 lsodesmosine .01 .46 .58 • 75

(a) Paper chranatography in n-butanol/acetic acid/water 4:1:5. (b) Paper chromatography in pyridine/acetic acid/water 30:20:50. (c) High voltage electrophoresis in 8\ formic acid.

-41- PAPER CHROMATOGRAPHY AND HIGH VOLTAGE ELECTROPHORESIS OF ELASTIN HYDROLYZATES. a AND b ARE GIVEN As Rf VALUES WHILE c AND d ARE REPRESENTED AS RATIOS OF LYSINE MOBILITY.

0 1.15 0 1.08 1.02 1.00 D 0 .95 0

.83 0 .Bl 0 0 .75 *o .74 0 .66 0 .66 0 .62 0 .60 .59 0 .54 0 .52 ~ .SI * 0 .46 . o .46 0 .41

0 .28 0 .22

. 0 .10

• ~~OS.01

0 0 0 a b c d

n-butanol/acetic pyridine/acetic pyridine/acetic 8% fonnic acid/water 4:1:5 acid/water 30:20:50 acid/water 1:10:89 acid HVE HVE

* indicates desmosines

Figure 22

-42- COLORIMETRIC AND OPTICAL PROPERTIES OF THE DESMOSINES AND THEIR ANALOGS

Employing nicotamide and N(l)methyl nicotinamide as analogs of the desmosines, several techniques for the detection of the pyridinium ring were pursued. The formation of the highly colored polymethine dyes from glutaconic aldehydes and primary aromatic amines afforded a specific and sensitive method for the detection of the nitrogen heterocyclic compounds. Figure 23 shows the formation of the glutaconic aldehyde after increasing the valence on the ring nitrogen from 3 to S with the addition of cyanogen bromide. This procedure permitted further investigation of the parameters for the formation of the polymethine dyes. Figure 24 illustrates the proposed reaction of desmosine with a primary aromatic amine. After cleavage of the quaternary ring with base, the amine would react with the resulting glutaconic aldehyde. Isodesmosine would be unable to participate in this process because of the ring substitution alpha to the nitrogen.

The NaBH4 reduction of N(l)methyl nicotinamide and isodesmosine was accomplished resulting in a change in their absorption spectra. The changes are significant enough to characterize these compounds.

REACTION OF CYANOGEN BROMIDE WITH NICOTINAMIDE Having established the absorption maximl.UJ\ of the cyanogen-nicotinamide

complex at 320 run, a series of reactions were run to detennine the temper~- ture and time required for CNBr to react with nicotinamide completely. To 1 ml of 1.056 mM- nicotinamide was added 1 ml of 100 rnM- cyanogen bromide at 24°C, 37°C, and 85°C. Absorbance readings were taken at 3 min-

-43- ute intervals against a nicotinamide-water blank. Figure 25 illustrates the course of the reaction at various temperatures. From this it was decided to run all subsequent reactions at 37°C for 20 minutes. pH OPTIMUM FOR REACTIONS OF PRIMARY AROMATIC AMINES WITII THE GLUTACONIC ALDEHYDE Three aromatic amines, benzidine ( E.•lf-aminodiphenyl ), E.-aminophenol, and E.-phenylenediamine were chosen as chromophores necessary for polymethine dye fonnation. (table 5).

Into 5 test tubes, 0.5 ml of 1.056 ~ nicotinamide and 0.5 ml of 100

~cyanogen bromide were added and incubated at 37°C for 20 minutes. At that time, 0.5 ml of 15.6 mt!.E.-aminophenol solution and o.s ml of various diluants were added to the test tubes. Similiarly, 0.5 ml of 3.275 m!:!_ nicotinamide and 0.5 ml of 100 nt.! cyanogen bromide were incubated as a bove. This time, 15.6 im-J.£_-phenylenediamine dihydrochloride and 15.6 m!:!_ E.•Ji-aminodiphenyl dihydrochloride were added separately to each of 5 test tubes. The 5 diluents were then added to the respective tubes. Table 6 illustrates the wavelength of maximum absorption of each of the dyes and the maximum absorbance at several pH values.

COMPARISON OF SENSITIVITY AMONG THE PRIMARY AROMATIC AMINES

Concentrations ranging from 3.275 ~ nicotinamide (l.637 pmole/0.5 ml) down to 0.0819 ~ nicotinamide (40.93 nnole/0.5 ml) were allowed to react with 0.5 ml cyanogen bromide (100 ~ as outlined above. The three pri mary aromatic amines were then added at a concentration of 15.6 IW.! each. Figure 26 shows the effects of both the aromatic ring and the number of amine groups on the sensitivity of the reaction when diluted with

-44- TABLE S

E.LI£-aminodiphenyl H2N-o-v-NH2

p-phenylenediamine

p-aminophenol

-45- H 0 /0 I c~ C" /,; c, CNBR H2o HC' CH O" 'NH2 'NH2 ) I II ... NH2CN + HBR O" HCOH O=CH

NC/ ' B\R Nicotinamide I Glutaconic aldehyde .;.. °'I 260nm 320nm

Figure 23 R R I I (C,H2)3 R (CH2>3 R R-(CH2l2 -0- (CH2l2R _ ____,. '(CHzlz '( <':, fACH2 ); _X---t ll_~~ O=CH HCO- N·a (C,H2)4 Alkali Enolate of

I ~ R Glutaconic Aldehyde -...J I Desmosine + H N- (CH ) - R 2 2 4 Lysine

X = p- aminodiphenyl

Figure 24 24°C 1.50

e ··-· • - • - • - • - • • 37°C 0 N I') / ·'-- 1.00 • / G> • au I ~ 0 I ----- 85°C Ill • ~ 0.50

10 20 30 40 Time (min) Figure 25 Reaction of cyanogen bromide(at three temperatures) with nicotinamide.

-48- TABLE 6 OPTIMUM pH FOR AROMATIC AMINES FO&~ING POLYMETHINE DYES

, . ABSORBANCE E.,p-am1no £_-phenylene- £.-amino- DILUENT oiphenyl di amine phenol 425 nm 415 nm 415 nm

2.0 N HCl 1.290 1.170 0.265 0.01 N HCl 0.840 0.930 0.285 pH 2.00 water 0.570 0.790 0.280 dilute NaOH 0.320 0.690 0.225 pH 11.85 2.0 -M NaOH turbid opaque

-49- 1.50 425 nm

Q) 1.00 a(J -e 0 ti) ~ o.so

so 100 150 200 x 10-B moles nicotinamide/2 ml

Figure 26 Comparison of sensitivity among l!.,!f.aminodiphenyl

( ) , E_-phenylenediamine (• • - • •) , and £_-amino-

phenol (• - -9 •

-50- 0,S ml of 2,0 -N HCl.

ABSORPTION SPECTRA OF POLYMETHINE DYE FORMATION

After allowing the cyanogen bromide and nicotinamide (3.275 rnM) to react as prescribed above, 0,5 ml of 15.6 ~E-phenylenediarnine dihydro chloride were added along with 0.5 ml of 2.0 !:!,HCl. 111e absorption spectra of the resulting colored solution is represented in figure 27. 111e same reaction was carried out for ..E.-aminodiphenyl at the same concentrations, and its absorption spectra shown in figure 28.

SODIUM BOROHYDRIDE REDUCTION OF N (l)METHYL NICOTINAMIDE

To test tubes containing 10 ml of 0.01628 Jn!iN(l)methyl nicotinamide iodide in 1.0 W.J. EDTA pH 9.3, 0.0526 to 3.170 mrnoles NaBH4 were added. lbe tubes were stoppered and thoroughly shaken in a Vortex mixer for several seconds. Aliquots of 2 ml were taken from each tube and their absorbance at 264 nm continuously monitored for 20 minutes. Table 7 represents the effects of both temperature and concentration of NaBH4 on the reduc tion of the pyridinium ring. Upon addition of sodium borohydride, the solution becomes very yellow. 111is color fades within several minutes with its absorbance at 358 nm decreasing parallel to that at 264 nm. Figure 29 illustrates the absorption spectra of both the reduced and non-reduced forms of N(l)methyl nicotinamide. 111is was obtained after reacting 1.628 umole of N(l)methyl.nicotinamide with 3.170 mrnole NaBH4 at 24°C for 60 minutes. Although N(l)methyl nicotinamide does not exhibit any fluorescence, the sodium borohydride reduced form was fluorescent when observed in the dark with a portable u.v. lamp.

-51- TABLE 7 REDUCTION OF N(l)ME1HYL NICOTINAMIDE WITH SODIUM BOROHYDRIDE

K>LES NaBH *absorbance change at 264 nm temperature°C 4

5.260 x io-5 0.095 24 1.585 x 10-4 0.120 24 1.585 x 10-3 0.285 24 1.585 x 10-3 0.245 37

3.170 x 10-3 o.38o** 24

* Readings are for 20 minutes. ** An additional 0.260 decrease was noted for the next 100 minutes.

-52- \ * Spectra obtained after 20 minute reaction of p-phenylenediamine.

I U1 ~ I

280 320 360 400 440 480 run Figure 27 Absorption spectra of p-phenylenediamine and cyanogen bromide complex. 0.60

o.so

0.40

I VI ~ I 0.30

0.20

280 320 360 400 440 480 1111 Figure 28 Absorption spectra of p-aminodiphenyl and cyanogen bromide complex. I I I non-reduced 0.10 . I I I I 0.60 I I I o.so I I I I 0.40 ' \ ,. G> ' u \ / \ ~ .., \ -e0 ti) 0.30 \ reduced ~ ,. ...-. \ \ , / ' \ \ I \ 0.20 \.' 1' \ ' I \ '-- \ ' \ \ 0.10. \ , __ _ ' '

280 320 360 400 440 nm Figure 29 Absorption spectra of reduced (----) and non-reduced ( ) N(l)methyl nicotinamide iodide,

-55- SODIUM BOROHYDRIDE REDUCTION OF ISODESHOSINE

A ~mall amount of isodesmosine (quantity undetermined) was dissolved

in 1.0 ~ EDTA pH 9.3. A 2 ml portion was mixed with 0.6341 mmole (24 mg) of NaBH and continuously monitored at 269 nm. In 30 minutes, the absorb 4 ance had decreased by 41% and did not change significantly thereafter. Figure 30 illustrates the absorption spectra of isodesmosine in EDTA before and after redution. Similiar to N(l)methyl nicotinamide, the reduced form exhibited strong fluorescence with a u.v. lamp. There was no noticeable increase in absorption between 300-360 nm for isodesmosine as there had been for the N(l)methylnicotinamide.

REDUCTION OF N(l)METHYL NICOTINAMIDE WITH SODIUM DITHIONITE

To 10 ml of 0.1628 ~N(l)methyl nicotinamide in 1.0 mt!_ EDTA pH 9.3 was added both 0,063 mmole (11 mg) and 1,896 mmole (330 mg) Na s o • 2 2 4 After reacting for 20 minutes at 24°C, the solution containing 11 mg show ed no change in absorbance. The solution containing 330 mg showed an in crease in absorbance at 264 nm.

-56- 0.60

o.so

0.40

\ non-reduced \ 0.30 \ \ \ \ \ 0.20 ' ' \

0.10 -...... - ...... ______

260 300 360 400 440 Jllll Figure 30 Absorption spectra of reduced (----) and non-reduced ( ) isodesmosine.

-57- DISCUSSION

Fractionation of elastin hydrolyzates by molecular exclusion followed by ion exchang~ chromatography can provide pure fractions of desmosine and isodesmosine(7). Although initial separation on Sephadex G-15 and Bio-Gel P-2 was accomplished with relative ease, many ion exchange tech niques were employed before an adequate resolution of these amino acids was achieved. Even though there appeared little difference in the elution pattern of amino acids between the Sephadex and Bio-Gel columns, the transition to the latter allowed larger quantities of hydrolyzates to be fraction ated. It was also demonstrated that the larger the Bio-Gel colunn, the better the separation of amino acids from the desmosines. Using 4% crosslinked sulphonated polystyrene resin (Dowex SOW X4) varying in particle size from 50-200 mesh (297-74 p diameter), large volumes of effluent were collected from various columns without obtain ing the desmosines. Only after employing these same resins with smaller particle size (19-47 p diameter) were adequate separations achieved. Hamilton(52) maintains that the uniform particle size of resins in amino acid chromatography is of great importance. He suggests that better resolution is obtained with resins of a smaller diameter due to rapid attainment of equilibrium between stationary and mobile phases. Fmploying resins of smaller particle size was not sufficient in it self. Alterations in column height, temperature, and buffer concentra tion and pH were necessary before an efficient means of resolving the desmosines was attained. The column with resin QlSS was obviously

-58- lacking the proper height to adequately separate the components of the elastin hydrolyzates. Columns employing Aminex MS "C" (fraction c ref. 8) resin yielded.adequate resolution of desmosine and isodesmosine only af ter the proper combination of these parameters was fol.Dld. The separation achieved, as shown in figure 12, was analyzed thor oughly by high voltage electrophoresis and automated amino acid analysis (PA-35). The two isomeric amino acids were desalted according to Anwar (13) and used as standards. Although several methods of paper chromatography and high voltage electrophoresis have been used, they were less adequate than the amino acid analyzer for identifying the components of the elastin hydrolyzates. Precluding their u.v. absorption spectra, this is the only method available to date which can differentiate between desmosine and isodesmosine. It is also considerably more accurate when working with samples containing a large variance in the concentration of amino acids. When a concentrated sample or one rich in desmosine and isodesmosine is available however, the high voltage electrophoresis method employing 8% fonnic acid has proven valuable. The sensitivity of detecting the pyridiniurn ring of a nicotinamide cyanogen bromide complex through the fonnation of a polymethine dye has been established with several primary aromatic amines. The color intensity of the various amines depends on pH, the number of amine groups, and the number of aromatic groups belonging to the chranophore. Benzidine(E.•E.! diaminodiphenyl) accordingly produces the most intense color yield of the three amines tested. The optimum color yield was also produced in 2.0 N HCl. Nicotinamide, after formation of the glutaconic aldehyde, reacted with

-59- benzidine to give a linear response at 425 nm, from 40-1600 nmoles/2 ml solution. Since the desmosines already exist in the quaternary fonn, the cleavage of t~e ·ring to form glutaconic aldehyde should provide a sensitive method for their detection. Both N(l)methyl nicotinamide and isodesmosine demonstrate changes in ultraviolet absorption spectra when reduced with sodium borohydride. Re duced isodesmosine loses its absorption at 269 nm, while the absorbance of N(l)methyl nicotinamide decreases at 264 nm with a simultaneous in crease in absobance at 358 nm. In agreement with Brignell(53), the re sulting dihydropyridines exhibit fluorescence. Either the loss of ab sorbance at 269 nm or the resulting fluorescence can be used to detect the desmosines more specifically than by their ultraviolet absorption at 274 nm(7).

-60- BIBLIOGRAPHY

1. Partridge,S.M., Elsden,D.F., Thomas,J., Constitution of the Cross-link ages in Elastin. Nature 197,1297-98(1963). 2. Thomas,J., Elsden,D.F., Partridge,S.M., Degradation Products From Elas tin. Nature 200,651-53(1963). 3. Seifter,s., Gallop,P.M., THE PROTEINS IV - COMPOSITION, STRUCTURE, AND FUNCTION by Hans Neurath, Academic Press, New York, 1966 pl53. 4. Partridge,S.M., Biosynthesis and Nature of Elastin Structures. Fed. Proc. 25,1023-29(1966) •. s. Gotte,L., Stern,P., Elsden,D.F., and Partridge,S.M., The Chemistry of the Connective Tissues - (8) The Composition of Elastin From Three Bovine Tissues. Biochem. J. 87,344-51(1963). 6. Franzblau,c., Sinex,F.M., and Faris,B., Isolation of an Unknown Com ponent from Hydrolyzates of Elastin. Nature 205,802-3(1965). 7. Ledvina, M., and Bartos,F., Detennination of Cross-links in Elastin. J. Chromatog. 31,56-61(1967). 8. Moore,s., Spackman,D.H., Stein,W.H., Chromatography of Amino Acids on Sulphonated Polystyrene Resins. Anal. Chem. 30,1185-90(1967). 9. Starcher,B.C., A Modified Two-column Procedure for the Analysis of the Basic Amino Acids Found in Elastin, Collagen and Antibiotics. J. Chromatog. 38,393-95(1968). 10. Corbin,K.W., Anal. Biochem. Ion-exchange Column Chromatography of Desmosine and Isodesmosine. 32,118-21(1969). 11. Yu,Y.s., Cross-linking of Elastin in Humin Atherosclerotic Aortas. Lab. Invest. 25,121-25(1971). 12. Gerber,G.E., Kemp,G.D., Amino Acid Analysis of Elastin--A rapid method. J. Chromatog. 71,361-62(1972). 13. Anwar,R.A., Comparison of Elastins From Various Sources. Can. J. Bio chem. 44,725-34(1966). 14. McConnick,D.B., Tu, s.c., Modified Procedure for Detennination of Small Quantities of Desmosine, Isodesrnosine, and Lysinonorleucine in Elastin. Anal. Biochem. 37,212-14(1970). -61- 15. Pickerell,J.A., Shafer,J., Lung Connective Tissue Measurements. Arch. Intern. Med. 127,891-95(1971}. 16. Ledvina,M. Bartos,F., Differences in the Cross-linking between High ly Purified Elastins from Bovine Elastic Tissues. Hoppe-Seyler's Z. Physiol. Chem. 351,661-62(1970). 17. Miller,E.J., Piez, K.A., An Accelerated Single-Column Procedure for the Automatic Analysis of Amino Acids in Collagen and Elastin Hydrolyzates. Anal. Biochem. 16,320-26(1966). 18. Thornhill,D.P., Separation of Amino Acids on Highly Cross-linked Poly acrylamide Gel and Some Observations on the Mechanism of Adsorption. Biochim. Biophys. Acta. 279,1-7(1972). 19. Thornhill,D.P., Desmosines: A Rapid Single-column Isolation Procedure. Anal. Biochem. 46,119-22(1972). 20. Moczar,E., Robert,B., andRobert,L., Separation and Detennination of Cross-linking Amino Acids of Elastin by High - Voltage Paper Electro phoresis. 45,422-47(1972). 21. Kivirikko,K.I., Laitinen,o., and Prockop,D.J. Modifications of a Spec ific Assay for Hydroxyproline in Urine. Anal. Biochem. 19, 249-55 (1967). 22. Prockop,D.J., Kivirikko,K.I., Relationship of Hydroxyproline Excretion in Urine to Collagen Metabolism-Biochemistry and Clinical Applica tions. Ann. Int. Med. 66,1243-66(1967).

23. Prockop,D.J., Undenfriend,S., A Specific Method for the Analysis of Hydroxyproline in Tissues and Urine. Anal. Biochem. 1,228-39(1960). 24. Geyer,G., Histochemistry of Elastin-Detection of Desmosines. Acta Histochem. 34,257-62(1969). 25. Glenner,G.G., Lillie,R.D., Observations on the Diazotization-Coupling Reaction for the Histochemical Demonstration of Tyrosine: Metal Chelation and Fonnazan Variants. J. Histochem. Cytochem.7,416-22 (1959). 26. Glenner,G.G., Nitrosophenol Reaction for Tyrosine and Related Compounds in Tissue Sections. J. Histochem. Cytochem. 7,423-24(1959). 27.. Lillie,R.D., Adaptation of the Morel-Sisley Protein Diazotization Pro cedure to the Histochemical Demonstration of Protein-Bound Tyrosine. J. Histochem. Cytochem. 5,528-32(1957). 28. Bandier,E., Hald,J., A Colorimetric Reaction for the Quantitative Esti mation of Nicotinic Acid. Biochem. J. 33,264-71(1939).

-62- 29. Fl'iedemann,T.E., Frazier,E.I.,The Use of "Tagged" Derivatives in the Fluorometric Assay of Vitamins. Science July 27th 97-99(1945).

30. Scudi,J.V., On the Fluorometric Detennination of Nicotinamide. Science io3,567-68(1946).

31. Feigl,F., SPOT TESTS IN ORGANIC ANALYSI~ Elsevier Publishing Company, New York (1960) pp281-2 & 290-91. 32. Anwar,R.A., Oda,G., Structure of Desmosine and Isodesrnosine. Nature 210,1254-55(1966). 33. Pullman,M.E., Pietro,A.S., Colowick,S.P., On the Structure of Reduced Diphosphopyridine Nucleotide. J. Biol. Chem. 206,129-41(1954). 34. Burton,M.R., Pietro,A.S., Kaplan,N.O., Preparation of Pyridine Nucleo tide Analogs by the Carbonyl Addition Reaction. J. Biol. Chem. Arch. Biochem. Biophys. 70, 87-106 (1957}. 35. Pullman,M.E., Colowick,S.P., Preparation of 2- and 6- pyridones of Nl-Methylnicotinamide. J. Biol. Chem. 206,121-27(1954). 36. Price ,J.M., The Detennination of N-Methyl-2-pyridone-5-carboxamide in Human Urine. J. Biol. Chem. 211,117-24(1954). 37. Lent,R., Franzblau,c., Studies on the Reduction of Bovine Elastin: Evidence for the Presence of ~6,7.Dehydrolysinonorleucine. Biochem. Biophys. Res. Corn. 26,43-50(1967). 38. Shimada,W. Bowman,A., Davis,N.R., Anwar,R.A., An Approach to the Study of the Structure of Desmosine and Isodesrnosine Containing Peptides isolated from the Elastase Digest of Elastin. Biochem. Biophys. Res. Com. 27,191-97(1969). 39. Starcher,B.c., Partridge,S.M., and Elsden,D.F., Isolation and Partial Characterization of a New Amino Acid from Reduced Elastin. Biochem. 6,2425-32(1967). 40. Lent,R.W., Smith,B., Salcedo,L.L., Faris,B., and Franzblau,C., Studies on the Reduction of Elastin. II. Evidence for the Presence of - 0. Aminoadipic Acidd'\-Semialdehyde and Its Aldol Condensation Product. Biochem, 8,2837-45(1969). 41.. Paz ,M.A., Gallop,P .M. ,Bltunenfeld,o.o., Henson,E., Seifter ,s., The Pres ence in Elastin of Possible Cyclic Precursors. Biochem. Biophys. Res. Cau. 43,289-97(1971). 42. Eisner, u., Kuthan,J., The Chemistry of the Dihydropyridines. Chem. Rev. 72,1-42(1972).

-63- 43. Lyle,R.E., Nelson,D.A., and Anderson,P. s., The Mechanism of the Re duction of Pyridinium Ions with Sodium Borohydride. Tetra. Let. 13,553-57(1962). 44. Yarmolinsky,M.B., Colowick,S.P., On the Mechanism of Pyridine Nucleo tide Reduction by Dithionite. BiochMl. Biophys. Acta. 20,177-89 (1956). 45. Lovesey,A.C., Ross,w.c., Potential Coenzyme Inhibitors. Part II. Re duction of 4-methylnicotinamide Derivatives by Sodium Dithionite and Sodium Borohydride. J. Chem. Soc. B 192-95(1969). 46. Smith, I., CHROMATOGRAPHIC AND ELECTROPHORETIC TECHNIQUES VOL II. ZONE ELECTROPHORESIS, Interscience Publishers (1968) pl76.

47 • Kremen,D.M., Vaughn,J .G., A Method for Single-Coll.mm Separation of Amino Acids. Fractions 2,8-10(1972). 48. Rubinstein,H.M., Pryce,J.D., The Colorimetric Estimation of Alpha Amino Nitrogen in Tissue Fluids. J. Clin. Path. 12,80-84(1959). 49. Anggard,E., Sjoquuist,B., Sjostrom,R., Chromatography of Biogenic Amine Hetaboli tes and Related Compounds on Lipophilic Sephadex The Phenolic Acids and Alcohols. J. Chromatog. 50,251-59(1970). SO. Williams,A.D., Freeman,D.E., Florsheim,W.H., Sephadex LH-20 Coltunn Separation of Thyroidal Iodoamino Acids. J. Chromatog. 45,371-80 (1969).

51. Joustra,M., Soderqvist, B., Fischer, L., Gel Filtration in Organic ·Solvents. 28,21-25(1967). 52. Hamilton,P.B., Ion Exchange Chromatography of Amino Acids - Effect of Resin Particle Size on Column Performance. Anal. Chem. 30,914- 19 (1958). 53. Brignell, P.J., Bullock,E., Eisner U., Gregory,B., Two Novel Rearange ment Reactions. J. Chern. Soc. 4819-27(1963).

-64- APPROVAL SHEET

lbe thesis submitted by Donald Louis Barbeau has been read and approved by three members of the faculty of Loyola University of Chicago. 1he final copies have been examined by the director of the thesis and the signature which appears below verifies the fact that any neces sary changes have been incorporated, and that the thesis is now given final approval with reference to content, form, and mechanical accuracy. The thesis is therefore accepted in partial fulfillment of the re quirements for the degree of Master of Science.

/O /4., ..: l I'??~ Date Signature of Adviser