METHOD FOR RAPID PURIFICATION, .ANALYSIS AND CHARACTERIZATION OF COLLECTIONS OF CHEMICAL COMPOUNDS

1. FIELD OF THE INVENTION The present invention relates generally to methods to manage the purification of a large number of closely related chemical compounds. More particularly the invention relates to the separation of mixtures which were deliberately created from a set of compounds with distinct chromatographic ,. retention times, distinct molecular weights, or other distinguishing features. In a preferred embodiment, a set of n compounds of distinct chromatographic retention times are combined, and then separated in a single chromatographic run, allowing a savings of time relative to the alternative of making n chromatographic runs. The present invention further relates to methods for separating compounds with at least one distinguishing feature by separations other than chromatography.

₂₀ 2. BACKGROUND OF THE INVENTION

Currently, there are many general methods of purifying synthetic compounds. These methods, which involve purification of a single target compound from multiple impurities, are undesirable for any multiple unit synthesis

25 because the chromatography of large numbers of compounds are difficult to chromatograph in series is very time-consuming and inefficient. This problem is multiplied in a many step synthesis; the requirement of purifying after each step consumes much time.

30 One example of a conventional technique for simultaneously synthesizing up to 36 individual peptides and thereafter screening said peptides is disclosed in an article by R. N. Zuckerman et al. entitled "Identification of highest- affinity ligands by affinity selection from equi olar peptide

35 mixtures generated by robotic synthesis," Proc. Nat. Acad. Sci., USA, Vol. 89, pp. 4505-09, May 1992.

For purification of an array, or a large number of compounds, the prior art teaches a repetitive method of chromatographing individual compounds. This constitutes a full cycle of synthesis, work-up, and purification for each molecule. While parallel synthesis has been a challenging issue, it is logistically far less complex and time consuming than parallel work-up and purification.

In general, there are several drawbacks with the teachings of conventional methods used by the prior art. In multistep procedures, there is a complexity not involved in single-step procedures, such that there is a greater possibility of error. Additionally, multistep procedures take more time and may use more reagents, and thus are more expensive. Further, the performance of these multiple chromatograms requires large amounts of solvent and time. The repetitive use and wear of columns and packings and subsequent loss of resolution is an additional consideration that could be remedied by consolidation of individual steps.

Therefore, it would be desirable to conduct chemical separations and purifications in a more efficient manner. The present invention provides various solutions to this problem.

3. SUMMARY OF THE INVENTION The present invention is based, in part, on the surprising discovery that concurrent coincidental chromatography of desired compounds is a practical, useful method for adequately purifying impure materials. The inventors observed that retention times of similar, but not identical, compounds possess baseline resolution or near- baseline resolution in HPLC chromatography. The present inventors also unexpectedly observed that a mixture of such crude individual compounds provided pure components upon simultaneous preparative chromatographic processing.

It is an object of the present invention to provide a method by which time, labor and expense is saved by combination of crude compounds for simultaneous purification.

Another object is to provide a method by which time, labor and expense is saved by combination of crude compounds and/or reaction mixtures for workup and simultaneous purification. Another object is to a method by which time, labor and expense is saved by a combination of crude mixtures, or even reactants for reaction, quenching, workup and simultaneous purification.

Yet another object of the present invention is to enable the synthesis of compounds in a large array format, previously thought to be impractical due to the presence of impurities, for the express purpose of generating large libraries for screening or other use.

Many other advantages are realized over the prior art with the present invention, for example, the practical capability for performing non-quantitative, average-yielding reactions in large numbers, due to the fact that now large numbers of compounds may be purified in parallel. Thus, separations which require only one step can be conducted with a significant savings in time. The current method teaches simultaneous chromatography of created mixtures and allows the processing of a much larger number of compounds with the same equipment and unit operations.

The present invention may be more fully understood by reference to the following detailed description, examples of specific embodiments and appended figures which are offered for purposes of illustration only and not by way of limitation.

4. BRIEF DESCRIPTION OF THE FIGURES

Figure 1 is a stacked composite of individual chromatograms of Compounds A-E as synthesized, prior to combination and purification.

Figure 2 is a preparative HPLC chromatogram of equimolar mixture of Compounds A-E.

Figure 3 is a stacked composite of individual chromatograms of Compounds A-E after simultaneous chromatography.

5. DETAILED DESCRIPTION OF THE INVENTION

5.1. Definitions

The following terms are used in the present invention to have the meanings set forth below:

"HPLC" encompasses "high performance liquid chromatography".

"Target molecule" refers to a molecule which is intended to be manipulated for the purposes of separation and/or purification. The target molecule, by definition, has no impurities. "Molecule of interest" refers to a target molecule in a phase which, prior to the application of a purification procedure, comprises both individual target molecules and at least one impurity molecule.

"Set of molecules" is a collection of related or unrelated molecules grouped for the purpose of eased manipulation, but not combined. Within a set of molecules, the user of the process defines which molecules are target molecules.

"Mixture" encompasses a set of molecules that has been combined deliberately for the sole purpose of expediting a unit operation or a chemical transformation.

"Purification" encompasses the eliminating of undesired contaminants and the recovery of desired material in an acceptable level of homogeneity. "Unit operation" encompasses any manipulation on a set of molecules as a whole, for example screening, work-up or purification.

"Selection parameter" refers to a physical property that is distinct for all target molecules in a set of molecules. One example of a selection parameter can be retention time in HPLC.

"Binding capability" refers to the capacity of a molecule or support to interact in a non-covalent fashion with a specified or unspecified ligand in a thermodynamically- favored process. "Compatible molecules" means molecules that do not chemically react with one another to form distinct chemical compounds.

5.2. Method for resolution of molecules According to one embodiment, the invention comprises a method for the resolution of molecules comprising four steps. First, one or more selection parameters are determined for each member of a set of molecules. A selection parameter could be retention time on HPLC. In this case, the HPLC retention time of each member of the set would be determined analytically or experimentally. Typically, for conventional chromatography equipment, analytical retention times with 15 to 90 seconds difference are preferable and the total operating window can be in the range of 15 to 150 minutes. Second, the molecules are selected on the basis of chemical compatibility and difference in at least one selection parameter. Chemical compatibility requires that selected molecules not react with one another to form distinct compounds. Difference in selection parameters requires that the selected molecules be distinguishable by experimental methods within the ability of one of ordinary skill in the art. For example, if HPLC retention time were the selection parameter, then the selection process requires each selected target molecule have a distinct HPLC retention time. Third, the selected molecules of interest would be combined together in any suitable manner, such as by simple mixing.

Fourth, the molecules of interest would be fractionated, such that the impurity level in the molecules of interest is lowered as a result of application of the method. The level of purity of a molecule of interest is somewhat arbitrary and dependent the particular application. Typically

Typically purities for one or more of these compounds of about 65% or more are required, and purities greater than 80-90% are often desired. Use of this invention can easily achieve these purities. This embodiment of the present invention is based on the discovery that, while it is counterintuitive to ever decrease the purity of a molecule in the course of purification, it is convenient to combine, then simultaneously purify, easily separable compounds. Therefore, combining compatible, crude molecules of interest, then chromatographing all of them simultaneously results in a substantial savings in time and material.

The selection of mixtures for the present method of purification via reversible absorptive interactions with the surface of a chromatographic stationary phase is based on knowing the retention time (RT) of every molecule to be purified in a given chromatographic system. The retention times can be sorted by time and combined so as to effect the an efficient throughput for the purification of the set of molecules. A molecule of interest can therefore be combined with other molecules of interest in the set under a given set of selection parameters. Examples of selection parameters can be chromatographic column load, minimum retention time difference, amount of a given molecule, and the maximum underlying impurity allowed to co-elute with a molecule of interest. L. R. Snyder and J. J. Kirkland, Introduction to Modern Liquid Chromatography. 1st ed., Wiley-Interscience, New York, 1974.

For example, each molecule of interest can be associated with a given chromatographic peak at a given retention time. Each time a new molecule of interest is proposed to be mixed with another molecule of interest or with a sub-set of the set of molecules to afford a new mixture, the following must be applied: #1. The selection parameter of the new molecule of interest must be compared to the selection parameters of the other

O 96/31263

molecules of interest already in the sub-set. There must not be a match.

#2. The selection parameter of the new molecule of interest must be compared to the selection parameters of the impurity molecules within the all the molecules of interest. There should not be a match. However, if the purity of a molecule of interest after the purification is acceptable, there may be a match in selection parameter of an impurity molecule with the selection parameter of a target molecule.

5.3. Preferred embodiment: HPLC separation In a preferred embodiment, between 4 and 12 molecules are selected such that they have discrete retention times in an HPLC and are mixed. The mixture is chromatographed on the HPLC and those fractions not containing one of the molecules of interest are discarded. The molecules of interest are collected and resorted according to a prearranged algorithm in a suitable receptacle or tray, preferably a 96-well plate.

5.4. Highly preferred embodiment: HPLC separation

In a highly preferred embodiment, one has a method for time resolution of molecules of interest comprising six steps. First, one determines the retention time on an HPLC instrument for each molecule in a set of molecules. Second, one selects a compatible grouping of molecules to form a set of molecules by optimizing on the basis of retention time, resolvability, and column parameters. Third, one forms a mixture of the molecules of interest. Fourth, one chromatographs the mixture on the HPLC to provide separation and purification of the molecules of interest. Fifth, one collects the fractions containing the molecules of interest. Sixth, one verifies analytically the identity of each fraction.

6. EXAMPLES In order to more fully illustrate the nature of the invention and the manner of practicing the same, the following examples are provided, which are not to be construed as limiting the remainder of the disclosure or the scope of the invention in any way whatsoever.

6.1

According to the present invention, a purification of five molecules of interest was done. The compounds, 4- benzylidene-1-(2-hydroxyethyl)-2-phenyl-5(4H)-imidazolone (denoted A), 4-(3-methoxybenzylidene)-l-phenylmethyl-2-phenyl- 5(4H)-imidazolone (denoted B) , 4-(3-methoxybenzylidene)-l-(2- (3-methylphenyl)ethyl)-2-phenyl-5(4H)-imidazolone (denoted C) , 4-(3-chlorobenzylidene)-1-(2-methylpropyl)-2-phenyl-5(4H)- imidazolone (denoted D) and l-(2-methylpropyl)-4-(4- phenylbenzylidene)-2-phenyl-5(4H)-imadazolone (denoted E) were synthesized by thermal rearrangement from the corresponding arylidene diamides, which were in turn derived from the sequential reaction of an aryl aldehyde with 2-phenyl-(4H)- oxazol-5-one, followed by ring opening with the corresponding amine. The molecular structure of the compounds are illustrated below. These five molecules are the target molecules of this example.

B

15

30

35

Purity of A-E as synthesized ranged from 35-57% by HPLC analysis, as shown below in Tables 1-5.

Table 1

Peak Number Retention Time Area Percent

1 3.08 1.30

2 3.14 2.18

3 7.29 3.17

4 8.69 3.49

5 8.90 22.41

6 10.94 2.89

7 11.90 56.80

8 14.04 5.14

9 17.15 2.63

Table 2

Peak Number Retention Time Area Percent

1 3.21 0.85

2 9.27 2.03

3 9.99 0.48

4 10.77 0.78

5 11.41 1.45

6 12.25 0.37

7 12.85 0.37

8 13.15 0.39

9 13.87 1.68

10 14.06 0.35

11 14.91 1.16

12 15.86 26.81

13 16.38 1.26

14 16.52 2.93

15 16.79 0.96

16 17.0 0.81

17 17.98 48.82

18 18.38 3.78

19 18.96 1.04

20 19.14 2.67

21 19.74 0.61

22 20.54 0.41

Table 3

Pe^ Number Retention Time Area Percent

1 9.94 0.51

2 10.33 2.87

3 11.51 0.86

4 12.55 1.79

5 14.69 1.96

6 14.95 0.64

7 15.75 1.19

8 16.37 0.62

9 16.70 26.53

10 17.04 0.51

11 17.18 1.94

12 17.57 0.49

13 18.05 4.03

14 18.27 0.42

15 18.60 0.46

16 18.83 44.64

17 19.11 0.98

18 19.72 5.02

19 19.95 1.05

20 20.13 1.56

21 20.33 1.00

22 20.76 0.55

23 21.49 0.40

Table 4

Peak Number Retention Time Area Percent

1 3.08 0.53

2 3.14 1.50

3 10.14 2.03

4 14.29 1.67

5 15.30 2.82

6 17.13 28.75

7 18.13 4.74

8 18.72 3.42

9 19.22 49.81

10 20.12 2.80

11 20.54 1.92

Table 5

Peak Number Retention Time Area Percent

1 3.20 0.64

2 7.13 1.31

3 9.52 0.43

4 11.51 1.01

5 11.71 0.74

6 12.15 2.36

7 12.39 0.25

8 12.74 0.18

9 13.37 0.20

10 14.14 0.40

11 14.68 0.36

12 14.84 0.48

13 15.15 0.48

14 15.32 0.80

15 15.77 5.81

16 16.12 0.28

17 16.31 0.83

18 16.84 0.17

19 17.03 0.49

20 17.43 0.29

21 17.82 19.18

22 18.20 0.45

23 18.41 0.51

24 18.79 1.03

25 19.09 5.46

26 19.26 1.06

27 19.40 1.24

28 19.57 0.27

29 19.93 1.23 30 20.19 35.04

31 20.34 2.63

32 20.64 1.13

33 21.01 0.72

34 21.27 0.48

35 21.45 0.91

36 21.56 0.88

37 21.80 8.71

38 22.34 0.70 39 22.66 0,86

A 20 mg sample of each crude mixture was combined in a holder of appropriate size and the five compounds were mixed.

This mixture was applied to a Zorbax ODS-SB RPC-18 column

(25 X 250 mm) and eluted with a 25 - 100% acetonitrile - 1% trifluoroacetic acid (TFA) :water gradient. Only those fractions corresponding to known HPLC retention times from previous analysis were collected. The order of elution was A,

B, C, D, then E at respective retention times of 25.7, 41.1,

43.1, 44.2 and 46.9 minutes. This* is shown in Table 6.

Table 6

Height Area

Peak No. Time Type (μV) (μV-5?c) Area

1 25.685 Nl 282531 23004896 19.798

2 27.640 N2 4484 97192 0.083

3 28.028 N3 4662 104349 0.089

4 41.126 N 363074 25180404 21.670

5 43.088 Nl 367402 23956860 20.617

6 44.191 N2 372506 22182110 19.090

7 46.871 N3 384929 21669772 18,649

•Total Area 116195583 99.996

Re-analysis of the separated compounds revealed purities of 85-97% by HPLC analysis and the identity was confirmed by mass spectrometry. This is shown in Tables 7-11.

Table 7

Peak Number Retention Time Area Percent

1 2.75 0.00

2 11.35 1.41

3 11.91 95.37

4 12.36 3.21

Table 8

Peak Number Retention Time Area Percent

1 2.76 0.00

2 15.70 1.05

3 17.49 0.99

4 17.90 96.77

5 18.29 1.19

Table 9

Peak Number Retention Time Area Percent

1 2.98 0.00

2 16.57 0.97

3 17.91 0.48

4 18.78 95.32

5 19.05 3.23

Table 10

Peak Number Retention Time Area Percent

1 2.72 0.00

2 15.64 1.33

3 17.10 0.55

4 18.79 1.56

5 19.19 84.50

6 19.37 4.60

7 19.70 7.47

Table 11

Peak Number Retention Time Area Percent

1 2.71 0.00

2 17.71 1.51

3 19.91 2.66

4 20.18 87.89

5 20.41 5.13

6 20.62 2.81

Graphic representations of the data are given in Figures 1-3. These figures represent the chromatograms and visually demonstrate the effectiveness of the present method.

These results demonstrate that the method by which molecules of interest are first combined and then purified is an effective method of separation.

6.2

The preceding example was repeated using another set of imidazolones consisting of 4-(4-ethylbenzylidene)-l-(2- hydroxyethyl)-2-phenyl-5(4H)-imidazolone (denoted F) , 4-(3- methylbenzylidene)-1-(methyl-(2-tetrahydrofuryl) )-2-phenyl- 5(4H)-imidazolone (denoted G) , 4-(3-methylbenzylidene)-l- phenylmethyl-2-phenyl-5(4H)-imidazolone (denoted H) , and 4-(4- ethylbenzylidene)-l-(1-naphthyl)-2-phenyl-5(4H)-imidazolone (denoted I) which varied in purity as synthesized from 47 - 76% by HPLC analysis. This data is shown in Tables 12 - 15.

Table 12

Peak Number Retention Time Area Percent

1 3.08 0.56

2 3.15 0.92

3 10.91 24.23

4 12.30 3.89

5 13.01 1.21

6 13.46 1.52

7 14.22 64.84

8 15.94 2.84

Table 13

Peak Number Retention Time Area Percent

1 3.08 0.44

2 3.15 1.26

3 4.16 3.36

4 6.23 1.25

5 8.61 1.64

6 10.50 1.19

7 13.48 25.49

8 16.58 63.33

9 16.99 2.04

Table 14

Peak Number Retention Time Area Percent

1 3.08 0.31

2 3.15 0.93

3 8.29 4.15

4 8.62 1.01

5 10.29 1.03

6 10.56 3.03

7 15.38 1.16

8 16.29 27.71

9 17.59 1.31

10 18.38 10.41

11 18.44 47.44

12 19.70 1.53

Table 15

Peak Number Retention Time Area Percent

1 2.70 0.02

2 10.35 9.80

3 12.69 2.67

4 17.93 1.58

5 18.15 1.43

6 18.49 3.65

7 19.25 1.32

8 19.71 1.25

9 19.96 0.98

10 20.36 76.32

11 21.40 0.98

Combination of a 20 mg sample of each and application to a Zorbax ODS-SB RPC-18 column (25 X 250 mm) , followed by elution with a 25 - 100% acetonitrile - 1% TFA:water gradient

afforded samples F- I in a purified form range from 95 - 99% in purity and eluting at 31. 2 , 37 . 6 , 42 . 1 and 47. 0 minutes . This data is shown in Tables 16-20.

Table 16

Height Area

Peak No. Time lϊES føY (μV-ξeς Area

1 21.690 N 5628 100973 0.093

23.423 N 646 8932 0.008

2 24.013 N 3715 73470 0.068

3 26.165 N 28671 1316715 1.224

4 28.116 Nl 12906 457401 0.425

5 28.751 N2 14267 36850 0.338

6 30.181 N3 16316 722004 0.671

7 31.238 N4 395451 28155410 26.179

8 33.510 N5 7827 176187 0.163

24.021 N6 909 2168 0.002

9 35.638 Nl 37569 1652023 1.536

10 37.556 N2 395539 30071292 27.960

11 40.036 N3 62577 3573319 3.322

12 41.118 N4 20645 215890 0.200

13 41.345 N5 21146 559845 0.520

14 42.126 N6 400632 22544290 20.92

15 44.320 N7 21642 572148 0.531

16 44.475 N8 22374 221193 0.205

17 45.068 N9 75664 2466974 2.293

18 45.976 N10 26719 1049299 0.975

19 47.033 Nil 406058 13051508 12.135

20 49.288 N12 5770 138795 0.129

21 49.885 N13 2640 54304 0,Q5Q

Total Area 107547990 99.989

Table 17

Peak Number Retention Time Area Percent

1 2.70 0.01

2 11.08 3.48

3 14.21 .96.51

Table 18

Peak Number Retention Time Area Percent

1 3.01 0.01

2 12.78 0.91

3 13.63 2.11

4 14.25 1.99

5 16.53 94.99

Table 19

Peak Number Retention Time Area Percent

1 2.92 0.00

2 16.33 2.64

3 17.78 0.36

4 18.42 96.00

Table 20

Peak Number Retention Time Area Percent

1 2.90 0.00

2 18.47 1.26

3 20.38 98.73