Jaynes- anti-inflammation
CERTIFICATE OF MAILING
I hereby certify that this correspondence is being deposited with the U.S. Postal Service Express Mail from Post Office to Addressee in an envelope addressed to Commissioner for Patents, P.O. Box 1450, Alexandria, 22313-1450 Alexandria on August 3rd 2010 with Express mailing number EH878150251 US.
Susanne Somersalo
PROVISIONAL PATENT APPLICATION
Inventor: Jesse Jaynes
Title: Anti-angiogenic peptides and their uses
TECHNICAL FIELD
This invention relates generally to the field of treating diseases. More particularly, the present invention relates to compositions and methods to prevent or treat angiogenesis and inflammation.
BACKGROUND ART
Angiogenesis is the process whereby new blood vessels are formed from pre-existing microvasculature (Fig. 1). This is a normal process that occurs during physiologic and reparative activities. Lack of proper spatial and temporal regulation leading to angiogenesis contributes to the disease process. Abnormal neo-vascularization is a predominant factor in diabetic retinopathy, arthritis, psoriasis, atherosclerotic plaques,

Jaynes- anti-inflammation
and cancer. It is important that inhibition of angiogenesis of endothelial cells is understood at the molecular level as it could lead to new ways of treating certain diseases.
Fig. 1 illustrates the angiogenic process. Blood vessel walls in arteries, arterioles, and capillaries, are lined by basement membrane composed of endothelial cells. Angiogenesis occurs mainly in the capillaries or post-capillary venules. In response to cytokine stimulation, endothelial cells break down the basement membrane, migrate into the extra vascular space, proliferate, and reorganize to form a new vessel. The endothelial cell carries its own internal defense against stray growth factors and it is the most sensitive of all cells to growth control by cell shape. With recent research, it seems that mechanical forces on a cell are necessary for growth factors, cytokines and hormones, to function. These soluble molecules will remain inactive unless they are coupled to the mechanical forces generated by specific insoluble molecules (collagen and fibronectin). These insoluble molecules lie in the extra cellular matrix and bind to specific receptors and integrins on the cell surface. This allows a cell to pull against its extra cellular matrix and to generate tension over the interconnected cytoskeletal linkages. Thus, cell shape changes are a prerequisite for entry of that cell into the cell cycle and subsequent gene expression and cell division. In fact, for the endothelial cell it is not the area and shape configuration of the outer cell membrane that supplies the direct mechanochemical information that permits DNA synthesis, but rather the shape of the nucleus. Nevertheless, nuclear shape is governed by the shape of the outer cell membrane and by tensile forces transmitted to the nucleus over the cytoskeletal network. When the shape of the nucleus is stretched beyond 60-70 microns there is net DNA synthesis.
It seems that its extra cellular matrix contains components; in particular, certain proteoglycans that bind and store these growth factors making them inaccessible to endothelial cells. For example, it is known that basic fibroblast growth factor (FGF) and vascular endothelial growth factor (VEGF) bind to heparin sulfate proteoglycan. The basement membrane itself may also inhibit endothelial growth. The laminin Bl chain contains two internal sites that, in the form of synthetic peptides having the sequence RGD and YSGR, inhibit angiogenesis. Furthermore, collagen XVIII is localized to the
Jaynes- anti-inflammation
perivascular region of large and small vessels and a 187 amino acid fragment, called endostatin, is a potent and specific inhibitor of endothelial proliferation. Several other endogenous proteins block the multiplication of endothelial cells and exert a reduced angiogenic effect. In each case, the endothelial inhibitor activity is found in a fragment of a larger protein which itself lacks inhibitory activity.
Angiogenesis has a role in development of various diseases. It is well known that angiogenesis is involved in development of malignant tumors and cancer diseases. Moreover, angiogenesis is associated with rheumatoid arthritis. Chronic inflammation may also involve pathological angiogenesis; examples of angiogenesis related inflammation diseases are ulcerative colitis and Crohn's disease. Chronic inflammation has been implicated to be the primary causative factor in several diseases including arthritis, multiple sclerosis, cervical spondylosis, tinnitus, systemic lupus, erythematosis, graft rejection, psoriasis, atherosclerosis, hypertension, and ischemia-reperfusion.
What is needed, therefore, is a composition and method that can effectively inhibit angiogenesis and chronic inflammation.
SUMMARY OF THE INVENTION
This disclosure demonstrates a physical connection between the structures of certain lytic peptides with known anti-angiogenesis factors. Chemical/structural similarities of chemocines involved in various diseases and certain lytic peptides are also shown. This disclosure demonstrates anti-angiongenic, anti-inflammatory and anti- cancer effects of certain lytic peptides.
BRIEF DESCRIPTION OF THE DRAWINGS
Fig. 1 illustrates the angiogenic process.
Jaynes- anti-inflammation
Fig. 2 provides physical characteristic of the 20 essential amino acids. The total volume, in cubic angstroms, is derived from the van der Waals' radii occupied by the amino acid when it is in a protein. Hydrophobicity is in kcal/mol and is the amount of energy necessary to place the amino acid, when in an alpha -helical protein, from the membrane interior to its exterior. Luminosity helps assigns the density of cyan (hydrophobic amino acids) or magenta (hydrophilic amino acids) to each glyph of the "molecular" font (Molly) that is described in this disclosure.
Fig. 3. Molly font wheel presented with single letter codes adjacent to each glyph. All hydrophobic amino acids are colored cyan while hydrophilic amino acids are magenta. The number values are relative hydrophobicities represented by the number of kcal/mole necessary to exteriorize an amino acid in an alpha helix from the inside of a lipid layer.
Fig. 4 illustrates three arrangements of naturally occurring peptides. The green band on the cylinders indicates the amino-terminus of the peptide while the gray band represents the carboxy-terminus. The cyan color represents regions that are predominately hydrophobic and the magenta color represents regions that are hydrophilic. Representative examples or natural peptides that fit this classification system are: mellitin-class 1; cecropins-class 2, and mangainins-class 3.
Fig. 5 shows sequences of natural lytic peptides melittin (SEQ ID NO: 104), Pipininl (SEQ ID NO: 105), adenoregulin (SEQ ID NO: 107), cecropin B (SEQ ID NO: 108), adropin (SEQ ID NO 110), magainin 2( SEQ ID NO: 110) and their optimized analogs JC1A21 (SEQ ID NO: 106), JC15 (SEQ IDNO:109) and JC3M1 (SEQ ID NO: 112) along with color scale representation.
Fig. 6 shows sequences of a defensin (SEQ ID NO: 113) and a protegrin (SEQ ID NO:l 14) along with an optimized analog JC41 (SEQ ID NO:l 15). Color scale representation is included.
Jaynes- anti-inflammation
Fig. 7 shows the sequence of Human Plasminogen protein (SEQ ID NO: 116) including the sequence of angiostatin protein (SEQ ID NO: 117) derived from it (underlined sequence). PL 1 (SEQ ID NO:l 18) and PL 2 (SEQ ID NO:l 19) are also shown with shadowing.
Fig. 8 shows sequences of fragments PL-1 (SEQ ID NO:l18) and PL-2 (SEQ ID NO: 119) derived from Human plasminogen protein. Color scale representation is included.
Fig. 9 shows the sequence of a fragment of Human Collagen XVIII (SEQ ID NO: 120). The underlined part of the sequence is the sequence of endostatin (SEQ ID NO:121). Fragment C-l (SEQ ID NO: 122) is shown with shadowing.
Fig. 10 shows sequence of the fragment C-l (SEQ ID NO: 122) derived from Human Collagen XVIII. Color scale representation is included.
Fig. 11 shows the sequence of platelet factor-4 (SEQ ID NO: 123). Shadowed sequences represent PF1 (SEQ ID NO: 124) and PF2 (SEQ ID NO: 125).
Fig. 12 shows sequences of fragments PF-1 (SEQ ID NO: 124) and PF-2 (SEQ ID NO: 125) derived form Platelet Factor 4. Color scale representation is included.
Fig. 13 illustrates Matrigel gels. A shows how a section of a Matrigel gel deposit looks like under the microscope soon after surgical implantation. The sample in B is derived from the control at the conclusion of the experiment. Intense activity is present with numerous cells attaching to the surface of the Matrigel. Cells begin to penetrate the deposit and organize into discrete structures that coalesce to form the beginning of tubes twisting and branching every way. In C, a typical sample from the peptide C-l treatment is shown. This treatment caused far fewer cellular associations evident at the perimeter of the Matrigel deposit. Consequently, there were far fewer cells and cellular structures inside of the Matrigel. Only one peptide fragment from JC15, JC15-10N, possessed anti-
Jaynes- anti-inflammation
angiogenic activity. A representative section of a Matrigel deposit from this set of animals is shown in D.
Fig. 14 shows anti-angiogenic activity of peptides of different lengths. As compared to control level the highest anti-angiogenic activity was obtained by peptides having less than 12 amino acids
Fig. 15 shows the sequences of natural and synthetic peptides of Example 6 in the color scale (Molly). The following sequences are shown : JC15 (SEQ ID NO; 126), JC15-18 (SEQ ID NO:52), JC15-15C (SEQ ID NO:53), JC15-10C (SEQ ID NO:91), JC15-12N (SEQ ID NO:67), JC15-10N (SEQ ID NO:54) C-l (SEQ ID NO:122). PF-2 (SEQ ID NO: 125), PF-1 (SEQ ID NO:124), PL-1 (SEQ ID NO:l 18), PL-2 (SEQ ID NO:l 19).
Fig. 16 illustrates the common motif of peptides of Example 6.
Fig. 17 shows the amino acid sequences of the chemokines of Table 7. The color scale is included and the sequences that are of interest are shadowed. Following chemokine sequences are shown: IL8 (SEQ ID NO: 127), MIG (SEQ ID NO:128), IP-10 (SEQ ID NO.-129), MCP1 (SEQ ID NO:130), MlP-la (SEQ ID NO:131), RANTES (SEQ ID NO: 132).
Fig. 18. Comparison of an endostatin fragment with full-length D2A21 peptide and its generated fragments displayed by Molly
Fig. 19A is a display of selected fragments from several cytokines and endostatin (derivation on the left, designation on the right) compared to 10N of D2A21. The light brown background illustrates conservation of hydrophobicity and resultant amphipathy. The dark brown background indicates those amino acids that are out of place.
Fig 19B displays the three dimensional representations of the peptide fragments obtained using the UCSF Chimera software.
Jaynes- anti-inflammation
Fig. 20. Changes in the absolute CD4+ Cell Counts in the sera of Nasty the lion before and after peptide treatment.
Fig. 21 Changes in the absolute CD8+ Cell Counts in the sera of Nasty the lion before and after peptide treatment.
Fig. 22 shows the weight profile and CD4+/CD8+ ratios of FIV-infected Hon before and after the peptide treatment. The CD4+/CD8+ ratios shown are absolute counts. The treatment consisted of weekly 70mg I.M injections.
Fig. 23 shows X-ray figures of a normal ankle, an arthritic ankle and an arthritic ankle after several peptide treatments (10 mg subcutaneous injections once a week for one month and then one injection per month for maintenance).
Fig. 24 is a graphical illustration of experiment in Example 12. The gray shades indicate endothelial cell growth flowing towards the presence of VEGF in the center of the rectangle Matrigel.
DETAILED DESCRIPTION OF THE INVENTION
To best illustrate the physical connections between proteins and peptides, it is necessary to display their sequences in ways that make it easier to visualize structural differences and similarities. There are a number of physical features that appear to be important in modulating the activity of the proteins:
1. Degree of amphipathy
2. Length of amphipathy
3. Heterogeneity of amphipathic section
4. Placement of amphipathic section (N or C terminal)
5. "+" Charge density (less or more)
Jaynes- anti-inflammation
6. Hydrophobicity of amphipathic section
7. Presence of hydrophobic tail
8. Length of hydrophobic tail
9. Hydrophobicity of tail
10. Placement of hydrophobic tail (N or C terminal)
11. Absence, presence, & position of "+" charged center
12. Absence or presence & position of flanking sequence
13. Predominating secondary structure
14. Termini modification (N-acetylation, C-amidation)
15. Surface area of hydrophilic and hydrophobic faces
16. Steric or volume considerations.
We can distinguish these characteristics by viewing the amino acids in ways that visually accentuate the differences in their physical attributes. In this respect, it is instructive to ponder the evolution of protein structure and the fact that, generally speaking, only 20 different amino acids are found in proteins. These are: alanine (A), arginine (R), aspargine (N), aspartic acid (D), cysteine (C), glutamic acid (E), glutamine (Q), glycine (G), histidine (H), isoleucine (I), leucine (L), lysine (K), methionine (M), phenylalanine (F), proline (P), serine (S), threonine (T), tyrosine (Y), valine (V), and tryptophan (W). There are a few exceptions, but these 20 are the only ones that are represented in the genetic code.
Taking these special 20 amino acids and viewing just two of their seemingly simple properties: hydrophobicity and volume differences can give one an appreciation of the significant chemical refinements that they must represent (Fig. 2.). The structural clues they provide in determining protein functionality are available, if we just look at them in the right way. For at least the last 2 billion years, life has found 20 amino acids, combined in different ways, to be adequate to meet all the challenges that it has faced on this planet. By applying combinatorial mathematics to the 20 amino acids, we derive, in practical terms, an almost infinite number of possible combinations that becomes an even
Jaynes- anti-inflammation
bigger number as the length of the protein is increased. For example, if we assume the maximum length for a protein is 200 amino acids, then, the total number of different proteins possible can be derived from the formula found below (sum of a finite geometric series):
(Formula Removed) "a1" is the first term, "n" is the number of terms, and "r" is the common ratio of the series increase, i.e., it goes up by a factor of 20 each time (the number of different protein amino acids). When one goes through the arithmetic, the number of possible combinations of proteins, from two amino acids in length to 200, is 8.458 x 10257. In addition, it should be noted that there are many proteins far larger than 200 amino acids in length. By studying the predominating 20 protein amino acids in certain ways, we can gain insight into the structural principles that govern all of protein biochemistry and then, as our awareness increases, subtle connections are discovered and seeming disparities can be replaced by recognizable physical commonalties.
In order to visualize differences and similarities in protein structure more easily, I designed a font (Molly) that is more representative of the chemical nature of the amino acids. To do this, I simply substituted circles with diameters equal to:

(Formula Removed)
for each particular amino acid. Then, setting the largest volume to 1, the smaller ones were proportionally reduced. Thus, the size of the circle is directly related to amino acid volume and, the differences shown between the amino acids, then, are visually accurate. To increase the information of the representation, the hydrophobicity or hydrophilicity of
Jaynes- anti-inflammation
each amino acid was converted to a color scale (Figs. 2 &3). The most hydrophobic amino acids are the most intense cyan color while those that are less hydrophobic are proportionally less concentrated cyan. Conversely, those amino acids that are most hydrophilic possess the deepest magenta color. Likewise, a graduated scale of less intense magenta color is used for those amino acids of lower hydrophilic character. Fig. 3 shows the color scale of the essential amino acids. From this scale (Fig. 2 and 3), it can be seen that, as amino acids become less hydrophobic or less hydrophilic, they become less pigmented and, therefore, more likely to be "exchangeable" within the protein structure. In addition, implicit in this scheme is that, within a particular hue, i.e., amongst hydrophobic amino acids or hydrophilic amino acids, of very similar properties, exchanges would be more likely to occur (generating the variability one observes in proteins of similar function from evolutionarily distant organisms). Of course, changes would be within the specific structural constraints imposed on each particular protein for it to retain its functionality. Most of the amino acid glyphs possess a mnemonic symbol that further characterizes its chemical properties. For example, charged amino acids have a "+" or a "-" sign incorporated within their glyph, the thickness of which, is related to the dissociation constant of their ionizable protons, other symbols aid in identifying the rest of the amino acids.
Lytic peptides are small basic proteins that appear to be major components of the antimicrobial defense systems of a number of animal species including insects, amphibians, and mammals. They consist of 23-39 amino acid peptides, which have potential for forming amphipathic a-helices or partial b-pleated sheets (locked by disulfide linkages); and thus, can interact with all cell types at the membrane surface. This interaction can result in no observable cellular effect, temporary cell impairment, death, or cell proliferation. That is why these molecules are more than lytic peptides.
Four distinct types of lytic peptides were discovered in the last decade; examples of each type are melittin, cecropins, magainins, and defensins. The properties of naturally occurring peptides suggest at least three distinct alpha -helical classes consisting of different arrangements of amphipathic and hydrophobic regions (Fig. 4). The green band
Jaynes- anti-inflammation
on the cylinders indicates the amino-terminus of the peptide while the gray band represents the carboxy-terminus. The cyan color represents regions that are predominately hydrophobic and the magenta color represent regions that are hydrophilic. Representative examples of natural peptides, which fit this classification system are: melittin-class 1, cecropins-class 2, and magainins-class 3 (note, 99% of all the known natural peptides fall within this classification system, data not shown). Therefore, separate synthetic peptides can be subdivided into distinct classes based on what has been observed in Nature.
Some examples of natural lytic peptides and their sequence as cast in the glyph motif are listed in Fig. 5, along with representative optimized analogs. These are shown in a typical linear array and are read from left to right.
The only natural lytic peptides that assume a b-conformation are the defensins and protegrins. They can assume this shape because of intra-disulfide linkages that lock them into this form, an absolute requisite for activity. We have completely novel classes of peptides that form b-sheets without the necessity of disulfide linkages. An example, JC41 is shown in Figure 6. The columnar array of hydrophobic and positive charged amino acids is apparent when the peptide adopts an amphipathic b-form. However, the width of the columns is narrower but overall length is greater than a peptide that adopts an amphipathic a-helix conformation.
Anti-angiogenesis
Lytic peptides are active in eliminating tumor-derived cells by causing direct osmotic lysis . Because of this demonstrable activity, it seemed logical that in order to demonstrate in vivo activity the peptide must be injected directly into the tumor. Indeed, that is the case. With just a few injections over a period of several days, tumors are permanently eliminated using the most active anti-tumor peptide, D2A21, yet tested. This begged the question: "What happens if this peptide is injected in a site removed from the tumor (in other words, can it express any systemic activity)?" The results were unexpected as most of the tumors also disappeared in several animal tumor models.
Jaynes- anti-inflammation
However, in some cases there was little activity. To determine what might be happening in vivo, radiolabeled D2A21 was chemically synthesized with all alanines labeled with either 3H or 14C. Since the labelling pattern was asymmetric, it enabled us to follow the physical state of the peptide once it had been injected into the animal by comparing the unique ratios of 3H/14C that would result if the peptide experienced proteolysis. It was found that within minutes the labelled peptide was hydrolyzed to fragments of various lengths no matter the route of administration but in the circulation approximately 14% of the radiolabel persisted for at least 24 hours with minimal further degradation (unpublished observations). The possibility emerged that the systemic in vivo anti-cancer activity was retained within specific fragments of D2A21. Based on these results, several peptide fragments were selected for further study.
Since similar responses are observed, from widely divergent disease conditions: bacterial infections and cancer, it is important to delineate the mechanism of action of the peptide fragments (and the cytokines/chemokines) as these seemingly different disease conditions must be fundamentally linked physiologically through the commonalities of protein/peptide structure/function. Supportive evidence has been obtained to conclude that one or more of the D2A21 fragments also modulate host-response to certain types of infection and one fragment, in particular, is acting as a potent anti-inflammatory agent.
The peptides of Table 1 are capable of reducing tumor mass when introduced from a "remote site", remote being defined as the administration of peptide in a place other than directly into the tumor.
100 Table 1. Amino acid sequences of various peptides. "L" denotes the length of the
peptide.
(Sequence removed)
Furthermore, metastatic disease was also significantly diminished (reduced by -90%) in the same experiments. Peptides of table 1 have an immediate effect on cancer cells, by causing a swift and lethal loss of osmotic integrity. Tumors, injected with these peptides, suffer immediate destruction that progresses, over a few days time, to eventual absorption by the surrounding tissue. At the cellular level, this is mediated by severe membrane perturbations as a result of physical contact with the peptides. Because of this well-established and observable direct killing activity, injection into tumors remains a viable treatment option, in some cases, particularly for head and neck cancer. However, if the peptides can function systemically as anti-tumor agents, their "usefulness" is greatly increased, and therefore, their value as a novel drug.
It is interesting to speculate about some of the peptides of Table 1, and the cell proliferative activity lytic peptides can promote, in the light of the recent findings about the prerequisite of cell shape changes for cell proliferation as discussed above. Consequential to peptide interaction with cell membranes is swelling of the cell and nucleus. Could peptide-induced proliferation simply be a result of their mechanical effects on the osmotic balance of the cell and by causing swelling, cell division occurs? Then, direct anticancer activity, the result of which might seem at first to be counterintuitive to proliferation, at least at the structural level, relies on the extreme disruptive nature of the peptides in altering the osmotic balance within the cancer cell. This activity overcomes any proliferative tendencies simply because a cancer cell and a normal cell differ in ways that can be exploited by the use of more active peptides.
Jaynes- anti-inflammation
Furthermore, a less lytic peptide would tend to shift the proliferation to kill equilibrium causing tumor growth, instead of destruction (Class 3 peptides).
The invention is now described by way of examples. These examples are not meant to limit the scope of the invention.
EXAMPLE 1. Selection of natural protein
With regard to so-called natural proteins, I focused on angiostatin and endostatin. Angiostatin (SEQ ID NO: 117) is derived from the plasminogen protein (SEQ ID NO:l 16) (see underlined region in Fig. 7). After comparing the sequence of plasminogen, with respect to important structural parameters for activity, I selected two fragments for testing (shadowed in Fig. 7). I call these fragments PL-1 (SEQ ID NO:l 18) and PL-2 (SEQ ID NO:l 19). Their sequence is shown in Fig. 8 with the color scheme. These two peptides were chosen from the larger sequence because they best fit my model regarding the postulated relationship of anti-angiogenesis activity to specific peptide sequences. What seemed to be important, in part, was the clustering of positive charge and amphipathy within a relatively small space on the peptide surface. This is assuming the protein adopts an a-helical structure in this region of its sequence (energetically speaking, an a-helix would be the most favorable conformation). The presumed function of plasminogen, once activated to plasmin, involves the dissolving of the fibrin of blood clots. It also acts as a proteolytic factor in a variety of processes including embryonic development, tissue remodeling, tumor invasion, and inflammation. Angiostatin has been shown to be a potent inhibitor of angiogenesis in vivo but at concentrations about 10 fold higher than what was needed to show remote anti-cancer activity with JC15.
Note that the selected fragments lie between two cysteines (Fig. 7). It is usually assumed that most of the cysteines in proteins will be involved in forming di-sulfide linkages that lock proteins into their preferred conformations. It seems to me that this can "present",
Jaynes- anti-inflammation
topologically speaking, the intervening amino acid sequences in a novel way to allow surface interactions.
Collagen XVIII's (SEQ ID NO: 120) fragment (SEQ ID NO:121) (underlined sequence in Fig. 9) is called endostatin. It, too, is a very powerful anti-angiogenic factor in animals. However, from a remote site, the peptides of Table 1 are more active than endostatin. In addition, some of them also directly kill the cancer cell, something that endostatin and angiostatin cannot do. However, if the short fragment selected from collagen XVIII was synthesized (shadowed in Fig. 9), it most probably would possess lytic activity. I call this fragment C-l (SEQ ID NO: 122) and its sequence is shown in Fig. 10 along with the color scheme. Note the clustering of charge, and other physical properties, is commonalties that permeate all of the selected fragments. Their overall structural "flavor" is so reminiscent of peptides of Table 1.
The last protein, from which I derived fragments, is platelet factor-4. It is ubiquitous in mammals and has been implicated in the angiogenic process. It also plays a major role in platelet aggregation and clot formation. I have found several regions in the protein that would possess a structure consistent with my model. Again, note that the regions appear between cysteine residues (shadowed in Fig. 11). I call these two peptides PF-1 (SEQ ID NO: 124) and PF-2 (SEQ ID NO: 125) and their sequences are shown in Fig. 12.
EXAMPLE 2. Selection of synthetic peptide sequences
In consideration of the above, the peptides listed in Table 2 were synthesized. Included are also several fragments of JC15. The possibility of bioactivity being retained in fragments of JC15 emerged because of data obtained from radioisotope labeling experiments that were conducted in rats (see Example 3 below). Basically, contrary to established dogma, JC15 is not immediately broken down to constituent amino acids, as has been found for other lytic peptides (for example, magainins are completely destroyed in minutes after injection). Rather, a significant portion (-14%) of JC15 remains in circulation for at least 24 hours after administration (see Example 3). But, it is not the full-length JC15 molecule; it is a mixture of fragments mostly between 10 to 18 amino
Jaynes- anti-inflammation
acids in length. Thus, the remote anti-cancer activity in animals must be due to one or more of its metabolized fragments, not the full-length molecule.
Table 2. Sequences that were synthesized and tested. MWT indicates the molecular
weight after addition of companion ions.
(Sequence removed)
Jaynes- anti-inflammation
EXAMPLE 3 Fate of JC15 when injected in animal
In these experiments JC15 peptide was labeled in the following way:
(Sequence removed) (SEQ ID NO: 126)
Where the shadowed italics denote the label is C14 and the underlined represents H3 labelled amino acids.
The specific activity is as shown below:
(Sequence removed)
(SEQIDNO:126)
29.3 1.46 39.8 41.4 56.3 1.96 mCi/mmole
Thus, we derive an expected H3 to C14 ratio of 48.86.
A stock of radiolabeled peptide was made by dissolving all of the labeled material (6.8 mg) in 680 ul of sterile water. This yields a solution that is 10 µg/ml in JC15 and 0.2235 µCi/ml (152 µCi/680 ul) in radiolabeled peptide. The dose for each animal (rat) is made by taking 9 ul of stock solution and adding enough cold JC15 to yield the final amount of 0.358 mg, all in a final volume of 200 ul. Several experiments were conducted using this stock of radiolabeled peptide. All tests followed the same basic protocol. Each of three rats was injected (by IP administration) with 200 ul of solution containing 0.358 mg of JC15 containing 2 uCi of radiolabeled peptide. One animal was sacrificed at each time point of 1, 6, or 24 hrs. After sacrifice, small tissue samples were dissolved with NCS solubilizer (1 ml/100 mg tissue) and decolorized with benzoyl peroxide in toluene (300 µ1/1 ml NCS solubilizer). After about two days incubation with agitation, 100 ul of the
Jaynes- anti-inflammation
solubilized tissue was added to 4 ml scintillation cocktail, incubated in the dark overnight and counted the following day. With regard to serum, 100 ul was taken and counted directly.
Table 3 shows the results of the experiments. It is interesting to note that even within one hour, there are significant counts appearing in the circulation. This level remains constant, even over the other two time points. Remarkably, after 24 hours, the number of counts is virtually the same as those found after 1 hr post-introduction of radiolabeled peptide. Interestingly, the other tissues (heart, lung, prostate, and injection site), at all time points, have less than 1% of the total radioactivity. Furthermore, the liver, which one would suppose would be the main site for the eventual proteolysis of the peptide, retains from 1.81% at 1 hr, 2.72% at 6 hrs, to just 2.11% at 24 hrs posMntroduction of the radiolabeled material. These results are unexpected as substantial amounts of peptide remain in the circulation even after one day after administration.
Table 3 Total counts and % of total of the radioactivity in different tissues after various time periods.
(Table Removed)
It is reasonable to assume that a significant amount of the peptide, once injected, is quickly bound to freely circulating constituents. Now, the important question to answer is "What is the physical state of the peptide after it enters the mammalian body"? Based on the seemingly rapid change in the isotopic ratio, once the peptide has been injected into the rat, it would seem that changes in the peptide, induced by host-proteolysis, have occurred.
I have derived a number of the possible degradative fragments of the radiolabeled JC15, based on the numbers above, including their expected H3/C14 ratios. It seems rather compelling that a significant amount of the peptide remains in circulation. With regard to the liver, there is a tendency for the ratio to become reduced below the original value. This is consistent with a generalized fragmentation whereby the fragments that are surviving longer contain, most probably, one each of the H3 and C14 labelled alanines. Of course, these counts reside within complex mixtures but the data do support the
Jaynes- anti-inflammation
generalized statement above. The converse is true when one observes the ratio of the counts in the serum. Indeed, it seems as if the structural integrity of the peptide is preserved to a greater extent in the sera, than in other areas of the body of the animal. This is borne out by the fact that the ratio increases with time in the serum. This is consistent with the supposition that there is maintenance of structural integrity with a gradual loss of amino acids, first, from the C-terminal end, followed by loss of several amino acids from the N-terminal region of the peptide.
Again, one must keep in mind the physical context of the peptide and assume that, within the animal, a very complex set of peptide fragments could be present. The following possible candidates were selected:
Peptide 1 is JC15, and its ratio as shown above is 48.77 (Sequence removed) (SEQIDNO:126)
Peptide 26-its ratio would be 114.25
(Sequence removed) (SEQIDNO:126)
Peptide 48-no ratio is possible to calculate
(Sequence removed) (SEQ ID NO: 126)
If we construct simple mixtures of these peptides, the ratios as in Table 5 apply.
Jaynes- anti-inflammation

Table 4.
(Table Removed)
Peptide mixtures and their ratios
(Table Removed)
As can be clearly seen from this example, the observed ratios, found in sera, over all time points, must contain peptide, in varying amounts, composed of the predominate species indicated (peptides 1, 26, & 48). These results support the hypothesis that a fragment of JC15 can be found that expresses the "remote" activity
EXAMPLE 5 Radioisotope labeling experiments with natural and synthesized peptide sequences
According to the procedure, Matrigel deposits were surgically implanted on both sides of 4 mice per treatment yielding a possible 8 samples per treatment. Matrigel is a polymeric substance that appears to be relatively inert in animals and it can serve as a matrix that allows experimentation in vivo on many different difficult-to-study-processes. Prior to
Jaynes- anti-inflammation
implantation, the Matrigel was allowed to imbibe fibroblast growth factor 1 (FGF1). This protein is a powerful inducer of angiogenesis and its presence guarantees that sufficient activity will be observed within the allotted time period of the experiment. Therefore, any inhibition of angiogenesis is likely to be a real phenomenon as the experiment has been set to heavily favor the angiogenic process. Angiogenesis occurs by day 14 and beginning Day 1 (Day 0 = day of implantation), mice were injected IP daily with 20 ug of peptide in 100 µl of normal saline. The animals were sacrificed on Day 14, and each Matrigel deposit divided longitudinally and fixed in 10% buffered formalin. One of the halves of each Matrigel deposit was then sectioned. Read-out for this experiment was via histology, with semi-quantitative/qualitative counting of migration of cells and their subsequent assembly of lumenal structures within the Matrigel. This method allows us to observe the full physiologic spectrum of effects, and was useful in delineating trends. Fig. 13 shows the summary rendition of what, on average was observed. In Figure 13, A represents what a section of a Matrigel gel deposit looks like under the microscope soon after surgical implantation. The sample in B is derived from the control at the conclusion of the experiment. Intense activity is present with numerous cells attaching to the surface of the Matrigel. Cells begin to penetrate the deposit and organize into discrete structures that coalesce to form the beginning of tubes twisting and branching in many directions. These venules eventually connect with the system carrying blood and it is possible to see red cells and lymphocytes within them. This process is called "arborization", derived from the fact that the angiogenic process most closely resembles the growth of roots and branches of trees. All but two of the peptide treatments looked, more or less, like B (all the peptides of Table 2 were tested). In C, a typical sample from the peptide C-l treatment is shown. This treatment caused far fewer cellular associations evident at the perimeter of the Matrigel deposit. Consequently, there were far fewer cells and cellular structures inside of the Matrigel. Only one peptide fragment from JC15, JC15-10N (SEQ ID NO:54), possessed anti-angiogenic activity. A representative section of a Matrigel deposit from this set of animals can be found in D. Importantly and surprisingly, not the numbers of internal cells and structures within the Matrigel deposit were reduced, but there was a seeming asymmetry of their organization where activity was evident. There were large regions of
Jaynes- anti-inflammation
Matrigel that had no visibly associated structures and few single cells, including on the periphery, while other regions had some limited activity. This experiment demonstrates that portions of endostatin and JC15 possess significant anti-angiogenic activity.
Table 5 shows the data collected from the experiment using semi-quantitative/qualitative scale for measuring angiogenic activity. This method is used as an initial assessment to find compounds that possess angiogenic activity, molecules that either accelerate or inhibit the process. This system ranks each sample using a 0 to 4 plus (+) scale. Thus, B in Figure 13 would yield a score of+++ while no + sign would yield a value of 0, as in A in Figure 13. In Table 3 the data is modified to a numerical form and plotted averages are shown.
TABLE 5 Data collected from experiments measuring angiogenic activity in semi-quantitative/qualitative scale.
(Table Removed)

Jaynes- anti-inflammation
(Table Removed)
Analysis shows that significant differences exist between the JC15-10N & C-l pair, from the rest of the treatments. However, JC15-ION (SEQ ID NO: 54) and C-l (SEQ ID NO: 122) are not significantly different from one another.
It is tempting to speculate that several of the peptides may actually promote angiogenesis. For example, mice treated with JC15-18N, JC15-15C, and JC15-12N, all show levels of activity higher than the control. Indeed, Matrigel deposits treated with the latter two peptides had the only top level (++++) scores of the entire experiment, a-amphipathic peptides of high positive charge density can cause cell proliferation, with the effect being more pronounced in peptides below 18 amino acids in length. These smaller peptides' lytic activity is greatly reduced because they are simply too short to physically span the membrane, the site of their direct mode of action. It is interesting that the level of activity is closer to the control in the full-length JC15 treatment group as opposed to some of its smaller fragments. A peak of angiogenic activity above the control is seen

Jaynes- anti-inflammation
when the peptide is between 12 and 18 amino acids in length, culminating in observable anti-angiogenic activity when the peptide is shorter than 12 amino acids (Fig. 14).
EXAMPLE 6. Structure/Function Relationships of the peptides and their Anti-angiogenesis effect
Table 6 allows one to see similarities or differences in the presence or absence of charged amino acids and their position with respect to hydrophobic (white rectangles) and other hydrophilic amino acids (dark rectangles) in the peptides tested in the Matrigel experiment.
TABLE 6 is a presentation of the peptides tested in the Matrigel experiment (Example 3) showing hydrophobic amino acids with white rectangles and hydrophilic amino acids as dark rectangles. See also Figs. 15 and 16.
(Figer Removed)
Jaynes- anti-inflammation
(Figer Removed)
One can see structural similarities, within sequence motifs, when sequences are presented as in Table 6. Of course, all of the fragments of JC15 are going to be identical to different regions of the full-length JC15 molecule. However, it is also apparent that the endostatin fragment, C-l, has more than just a passing resemblance to JC15 and its fragments, as do portions of the peptides from plasminogen. In addition, the C-terminal half of PF-2, derived from platelet factor 4, shares similarly significant structural homology.
Jaynes- anti-inflammation
As can be seen from Figure 19, there is a close physico-chemical relatedness of CI (EndoF) and JC1510N (D2A21-10N) when illustrated with Molly.
Are these structurally homologous regions enough alike to all modulate angiogenesis in some way? Most biochemical processes occur at the surfaces of different macromolecules that associate or bind to specific regions on one another within a discrete three-dimensional space. These binding sequences are often rather short stretches of a protein, say, 4 to 8 amino acids. It is entirely within the realm of possibility that there are only 5 or so amino acids that comprise the critical binding region that interacts specifically with target macromolecules initiating an in vivo anti-angiogenic response. The data support the hypothesis that C-l and JC15-10N possess this binding region.
In Fig. 15 the sequences are casted in Molly and Fig. 16 is a simple schematic illustration derived from Fig 15. By keeping in mind that each magenta square is, with just a few exceptions, a "+" charged amino acid, the following conclusions can be made:
• JC15 and all of its fragments possess the same type of internal sequence of 7 or 9 amino acids, with JC15 and JC15-18N retaining one of each. Noting the shift of one amino acid, most importantly, the same can be said for the peptides C-l, *PF-1, and *PF-2.
• The anti-angiogenic fragment must be of a certain length. Even if a fragment retains the putative 7 or 9 amino acid binding sequence, like JC15, JC15-18N, JC15-15C, and JC15-12N, it still cannot exert an anti-angiogenic effect. Clearly, the simplest explanation is that these sequences cannot "fit" into the target-binding site. How critical this size requirement is, can be borne out by the fact that a fragment identical to JC15-10N, but with the addition of two amino acids, JC15-12N, does not inhibit angiogenesis. In fact, it may actually cause an opposite effect. Then, one may ask, why does C-l possess anti-angiogenic activity when it seems to violate the size requirement, after all, it is 21 amino acids in length? My best guess, at this time, is that the proline, with just 2 amino acids separating it from the
Jaynes- anti-inflammation
putative binding sequence, directs the rest of the fragment away from the target-binding site, reducing interference to a minimum. After all, that is proline's function—to allow bends and turns in proteins. Alternatively, it could be processed in the animal to a shorter fragment.
• More than a specific length is necessary. JC15-10C and JC15-10N are the same size yet JC15-10N is the only one that possesses anti-angiogenic activity. Even though JC15-10C contains a probable 7 amino acid binding sequence, the addition of 3 hydrophobic amino acids on the C-terminal end of JC15-10C are enough to negate binding, 2 of the 3 being bulky phenylalanines. In addition, one can conclude that a more optimal binding fragment contains several pairs of charged or other hydrophilic amino acids in the binding sequence see JC15-10N and C-l. Perhaps, another reason why JC15-10C was inactive.
• The "interchangeability" of like amino acids is most apparent in comparison of JC15-10N with C-l. Even though their sequences are quite different, almost perfect correspondence is observed when they are cast in the molecular font. It is possible that JC15-10N could be made even more active by removing one of the internal hydrophobic amino acids and reducing its length by one or two amino acids from its C-terminal end. Also, the addition of a negatively charged amino acid, within the charged pair, may be desirable.
EXAMPLE 7. Chemokine Anatomy and the Design of Novel Domains to Delineate Specific Cellular Activities
Chronic inflammation has been implicated to be the primary causative factor in various diseases including: arthritis, multiple sclerosis, cervical spondylosis, tinnitus, systemic lupus, erythematosis, graft rejection, psoriasis, atherosclerosis, hypertension, and ischemia-reperfusion. The surprising fact is that just a handful of pro-inflammatory chemokines are responsible and according to this disclosure JC15-10N has structural analogies within the sequences of each molecule.
Jaynes- anti-inflammation
While there are more than 50 chemokines that have been characterized, but a clearly smaller set is involved in diseases. Table 7 provides internal sequence of a number of chemokines and Table 8 shows the chemokines involved in several diseases.
Jaynes- anti-inflammation
findostatin ■ F-Collagen XVIII IVRRADRAAV
Table 7 Amino acid sequences of selected domains derived from several cytokines, oncostatin and endostatin. (Figer Removed)
Table 8. Chmemokines involved in development of various diseases. The sequences of the chemokines are shown in the sequence listing with the following sequence numbers: IL8 (SEQ IDNO:127), MIG (SEQ ID NO:128), IP-10 (SEQ ID N0129), MCP1 (SEQ ID NO: 130), MlPla (SEQ ID NO: 131) and Rantes (SEQ ID NO: 132).
The chemical/structural similarities of the chemocines in Figure 19 A with JC15-10N are easy to recognize. They conserve amphipathy and charge density to a high degree and their 3-dimensional structure (Fig. 19B) would be quite similar to JC10. Mostly they all appear after a proline and are more often than not at the C-terminus—this yields distinct domains. Also, it is interesting that the internal sequence of WVQ has been conserved with the divergent one IP-10 possessing AIK that conserves hydrophobicity exactly. I would predict that all of the above sequences would possess anti-angiogenic and antiinflammatory activity much like JC15-10N. Thus, these key sequences of each domain, within the specific protein, no doubt functions as a down-regulator or off/brake switch for the inflammatory process.
Jaynes- anti-inflammation
EXAMPLE 8 Antiangiogenic and anti-inflammatory effects of JC15-10N as tested in a lion infected with FIV
Nasty is a male lion in North Carolina Zoological Park. He was diagnosed to suffer Feline Immunodeficiency virus FIV. FIV attacks the immune system of cats, much like the human immunodeficiency virus (HIV) attacks the immune system of human beings. FIV infects many cell types in its host, including CD4+ and CD8+ T lymphocytes, B lymphocytes, and macrophages. FIV eventually leads to debilitation of the immune system in its feline hosts by the infection and exhaustion of T-helper (CD4+) cells.
Nasty was treted weekly with 70 mg I.M injections of JC15-10N.
Figure 21 shows changes in the absolute CD4+ Cell Counts of Nasty before and after peptide treatment. It can be seen that starting of peptide treatment stabilized the CD4+ cell counts.
Figure 22 shows changes in the absolute CD8+ cell counts of Nasty before and after peptide treatment. Starting of the treatment prevented the decrease and actually, the cell counts began to rise soon after the treatment.
Figure 23 shows weight profile of Nasty before and after the peptide treatment along with changes in CD4+/CD8+ ratio. As can be seen, the weight of the lion began to rise immediately after beginning of the peptide treatment.
EXAMPLE 9 Treatment of Pancreatic Cancer with JC15-10N
A 73 year old woman diabetic since 12/01 was diagnosed with Stage IV pancreatic cancer in May 2002 with metastatic diseases in her liver. Median survival time of patients with Stage IV pancreatic cancer is 4.5 months. Median survival time of patients
Jaynes- anti-inflammation
with Stage IV pancreatic cancer when treated with gemcitabine is 4.8 months. The longest anyone lived on gemcitabine treatment has been 19 months.
The patient of this case started JC15-10N peptide treatment on August 2002. The patient was given 0.5 mg/kg peptide sub-cutaneously. The patient weighted 128 lbs and therefore she received 29mg/injection per week. The disease in her liver diminished significantly. The patient required no more diabetic medication, which indicated that her primary tumor was regressing. After receiving the peptide for more than 19 months, the patient was doing fine. The patient passed away from unrelated causes on 7/2004.
EXMAPLE 10 Treatment of arthritis with JC15-10N peptide
A patient with arthritis was treated by subcutaneous injection of 10mg once a week for one month and then once a month for maintenance doses. A visible indication of arthritis is calcification of joints. The calcification of the ankle joints disappeared during this time indicating The results are shown in Figure 24.
EXAMPLE 11 Treatment of murine modle of ulcerative colititis with JC15-10N in IL-10 deficient mice
Interleukin 10 (IL-10) is known for its anti-inflammatory properties in mammals. Several cell types including monocytes and lymphocytes produce it. It has been shown to down-regulate Th 1 cytokines, MHC class II antigens and co-stimulatory molecules on macrophages. There is good evidence that it also acts as an immuno-regulator in the intestinal tract and plays a positive role in limiting inflammatory bowel disease (IBD) in humans. Clinical research has demonstrated that patients with inflammatory bowel disease, ulcerative colitis and Crohn's disease are predisposed to cancers of the intestinal tract.
An IL-10 deficient strain of mouse was used in the study to determine the ability of C 15-10N to limit their developing IBD and subsequent colon cancer.
Jaynes- anti-inflammation
The Wild Type and IL-10 deficient mouse strains used in this study was obtained from Jackson Laboratories and are designated as:
129SvEv Wild Type
129SVEV-IL10-'-The following groups of animals comprised our study: Controls:
1. Controls (129 SvEv 129-/- untreated)
2. Controls (129 SvEv 129-/- Sham/Saline Only)
3. 129 SvEv Wild-type Controls (untreated)
4. 129 SvEv Wild-type (Sham/Saline Only)
Treatments with JC15 ION:
1. Pre-inflammatory (129 SvEv 129-/- -Prevention-1 injection/week)
2. Frank inflammatory (129 SvEv 129-/- -Treatment-1 injection/week)
3. Pre- inflammatory (129 SvEv 129-/- -Prevention-1* injections/week)
4. Frank-inflammatory (129 SvEv 129-/- -Treatment-1* injections/week)
5. 129 SvEv Wild-type (Prevention-1 injection/week)
6. 129 SvEv Wild-type (Treatment-1 injection/week)
7. 129 SvEv Wild-type (Prevention-1 * injections/week)
8. 129 SvEv Wild-type (Treatment-1* injections/week)
9 animals/group X 12 groups (including controls) =108 mice x 2 repetitions = 216 mice; 1 injection/week = 0.5 mg/kg subQ and 1* injection/week = 5.0 mg/kg subQ injection/week.
In order to encourage inflammation, prior to the start of the experiment, pathogenic murine strains of E. coli and E.faecalis bacteria were administered to the mice in the appropriate treatment groups via oral and anal gavages, designated EC/EF.
Jaynes- anti-inflammation
Pre- inflammatory (4 wks.) and frank- inflammatory animals (6 wks. - 8 wks) were treated with GP4 via subcutaneous injections every week for 14 weeks. The compound was diluted in saline to obtain treatment doses of 0.5 and 5.0 mg/kg of body weight respectively and was delivered subcutaneously using tuberculin needles. As controls, some animals were injected with saline only or remained uninjected during the course of the experiment. The procedure described by Hem and coworkers (Laboratory Animals Ltd. 1998. V 32. 364-368) was used to collect 100[xl of blood from the lateral saphenous vein of all test animals once per week over the 14 weeks of peptide therapy. Blood serum was analyzed for changes in the levels of pro-inflammatory and anti-inflammatory cytokines over the 14-week treatment period. At the end of the 14-week treatment period animals were euthanized using a CO2 chamber and flushing the peritoneum with PBS will collect peritoneal lavage fluid. Colons were removed and flushed separately with PBS. Cytokine levels in the peritoneal lavage fluids and colonic fluids of all animals were compared. Following colonic flushes; colons were splayed and examined for inflammatory lesions. Lesions were excised with scissors and the remainder of the colon rolled into gut rolls. Both lesions and gut rolls were fixed by incubation for 24 hrs. In 10% formalin, rinsed in 70% ethanol and held in PBS until they were embedded and prepared for histological analyses.
The influence of treatments on the numbers of circulating immune system cells including macrophages, neutrophils, dendritic cells and lymphocytes was monitored. Immune cells were isolated from the blood and peritoneal lavage fluid collected and their numbers quantitated. Additionally, primary (thymus and bone marrow) and secondary (spleen and lymph nodes) lymphoid tissues were removed from all animals at the time of sacrifice for comparative histological analyses. The sera and tissue samples of both are in the process of being analyzed.
Preliminary data analyses demonstrate a profound protective effect of JC15 10N at both dosages (with 0.5 mg/kg being somewhat better than the 5.0 mg/kg dosage). There was significant reduction of colitis, presence of advanced disease and subsequent development of colon cancer in JC15 10N -treated IL-10-/- mice compared to the control IL-10-/-

Jaynes- anti-inflammation
receiving no treatment. The wild-type mice showed no disease symptoms across all treatments.
EXAMPLE 12. Ability of 10 N to inhibit migration of endothelial cells
Table 10
(Table Removed)
Jaynes- anti-inflammation
The results in Table 10 are gathered from an in vitro experiment and show the ability of JC 15 10N to inhibit the migration of endothelial cells (the "minus" values). This migration is a requisite step in the formation ofnew blood vessels. The fact that 10N retards this activity demonstrates its ability to block angiogensis and hence explains, at least partially, its inhibitory effect on cancer. The shading indicates the presence of VEGF (vegetative endothelial growth factor). This growth factor causes endothelial cell migration and consequent assembly of blood vessels. Because 10N inhibits this migration even in the presence of VEGF (gray shading) is highly significant.
A graphical representation of this type of experiment is shown in Figure 24. The grey shades indicate endothelial cell growth flowing towards the presence of VEGF in the center of the rectangle of Matrigel. The retardation or enhancement of cell movement over the controls can be seen clearly.
Cell Migration Assay
Cell migration was performed as previously described. In brief, a HMEC-1 monolayer was scraped making a I-m wide denuded area then stimulated with VEGF and 10N and the area unoccupied by the migrating cells was determined using MetaMorph and expressed as a percentage of control.

We CLAIMS
What is claimed is:
1. A method to treat angiogenesis and inflammatory related diseases, by
treating apatient with at least one peptide sequence possessing physico-
chemical relatedness with SEQ ID NO: 54 (JC15-10N) when illustrated
with Molly.
2. All sequences shown in this disclosure should be claimed for anti
inflammatory and/or anti-angiogenic activity, including those that are
derived from other proteins.
3. Technique of using the claimed molecules to treat inflammatory diseases,
including but not limited to arthritis, rheumatoid arthritis, cancer, sepsis,
cardiovascular disease, cerebrovascular disease, diabetes, inflammatory
bowel disease, IIIV, sickle cell disease, cerebral malaria, cystic fibrosis
and Crolum's disease, etc.
4. Sequences relevant to promoting clot formation, i.e., the platelet factor 4 fragments.
5. Technique of using fragments to promote clot formation.
6. By inspection, other sequences in platelet factor 4 are selected as those
that inhibit clot formation (highlighted in yellow below).
EAEEIX5DLQCLCVKTTSQVRPRHITSLEVIKAGPHCPTAQLIATLK NGRK1CLDL ,QAPL ,YKKII KKLL,ES
7. Technique of using the fragments in (5 above) to inhibit clot formation.
8. Any peptide, peptoid, peptide mimetic sequence of the pattern below to
be used as a treatment molecule. Where blue circles represent amino
acids containing hydrophobic groups and magenta circles represent amino
acids containing hydrophilic groups.