WO1990014593A1 - Methods and compositions for the identification of patients sensitive to interferon alpha - Google Patents

Abstract

Disclosed are DNA segments and method useful in the identification of cancer patients who are sensitive to interferon alpha therapy, thereby providing a means for identifying those patients that can be successfully treated with interferon. Preferred DNA segments incorporate a Friedman-Stark Response Sequence (FSRS), an Interferon Sensitive Response Element (ISRE) and an Octomeric Region sequence (e.g., -T-G-T-C-G-T-C-A-). The assay involves a DNA/protein binding procedure using a protein fraction which includes nuclear proteins prepared from patient's blood cells pretreated with interferon alpha. In a gel mobility shift assay, patients sensitive to interferon therapy exhibit a distinctive Low Mobility Complex not exhibited by patients who are not sensitive.

Description

METHODS AND COMPOSITIONS FOR THE

IDENTIFICATION OF PATIENTS SENSITIVE

TO INTERFERON ALPHA

The government may own certain rights in the present invention pursuant to NIH grant ROl HL 29300.

The present invention relates to methods and

compositions useful in predicting the sensitivity of a cancer patient to treatment with interferon alpha. In particular aspects, the invention concerns DNA sequences and segments which can be employed in assays to predict the potential usefulness of interferon alpha therapy in cancer patients. Interferon alpha (IFN-alpha), also known as leukocyte interferon, is one of three distinct types of interferons that have been characterized, the others being "beta", or fibroblast interferon, and "gamma", or immune interferon. The interferons are all proteins that are naturally produced by the body, generally in response to some inducing event such as a viral infection or through the use of a variety of inducing agents. It was discovered early on, when it was believed that there was only one class of interferon, that this interferon possessed an antiviral action that could perhaps be taken advantage of in developing interferon as an antiviral agent.

Later experiments indicated that interferon was not the actual antiviral agent, but rather that it induced an "antiviral state" in cells which rendered the cells resistent to viral infection. It was postulated that interferon induced the cells to produce a certain protein or proteins that were themselves the actual effectors of virus multiplication. Several proteins that are induced by interferon treatment are now known. Whether these proteins are the actual inhibitors of virus multiplication is not yet clear (1). Major insight was gained in the study of the

interferons during the 1960s and 1970s. For example, it was discovered that there were in fact three different types of interferons, which prepared the way for the later discovery of families of interferon genes. A second major insight gained during this period was that interferon possesses not only antiviral activity, but also cell multiplication inhibitory activity and cell regulatory activity. This realization that interferons affected cell growth stimulated a great interest in the interferons as anticellular or antineoplastic agents.

More recently, the genes encoding each of the three classes of interferons have been cloned and expressed, giving medical science a relatively large supply of each of the various species for more directed research and even clinical trials. While the interferons have not proven to be particularly valuable in antiviral therapy, except perhaps in persistent viral diseases, they have shown some promise in antitumor therapy. Through extensive clinical trials using recombinantly produced interferons, the interferons have shown a varying degree of success in the treatment of a variety of tumors.

While the mechanism of action of IFN-alpha is unclear, it is known that IFN-alpha will activate the transcription of various immunomodulatory and antiviral genes. It is proposed that this specific gene activation is brought about by an alteration in the binding of nuclear proteins to the 5' transcription-regulatory regions of IFN-alpha inducible genes (2-5). It is

hypothesized that IFN-alpha can, is essence, reprogram the subset of genes being transcribed in a particular cell and can thereby directly or indirectly effect a reduction in tumor cell growth. For example, IFN-alpha has been shown to induce a reduction in the percentage of Philadelphia chromosome-positive cells and a regrowth of normal diploid hematopoietic cells in some CML patients in early chronic phase (6-10).

It has become apparent that one of the mechanisms by which interferon acts is by means of an interaction with high affinity plasma membrane receptors, through which it directly alters the expression of a number of genes in responsive cells (11). The protein products of interferon inducible genes belong to two major groups: 1) surface antigens, especially Class I and II HLA antigens and Beta- 2 microglobulin, and 2) enzymes which modify double stranded mRNA to render it unsuitable as a template for translation: double stranded RNA dependant protein kinase, 2,5 oligoadenylate synthetase, 2,5 oligoadenylate

activated nuclease, and 2' phosphodiesterase (22). The effect of interferon on the expression of these genes is uniformly at the level of transcription initiation and in some cases, additional effects on processing of mRNA degradation are seen.

Recently, a sequence common to interferon inducible genes (HLA as well as 2,5 oligoadenylate synthetase gene) has been identified (4) , which has been shown to be a binding site for nuclear trancriptional regulatory

proteins. These elements are 5' to the genes for the HLA antigens and to the translational inhibitory enzymes such as 2,5 oligoadenylate synthetase (13). These interferon response elements exhibit sequence homology with each other, and a consensus sequence for a possible DNA/protein recognition site has been proposed (4). The binding of nuclear proteins to DNA follow within minutes of the exposure of the cell to interferon, and follows the time course of activation of gene expression by interferon.

Deletion of the binding sites for these nuclear proteins abrogates the activation of transcription initiation of these genes (4). Gutterman and his colleagues previously reported that the evolution of interferon resistance in CML did not always involve loss of binding of interferon to its plasma membrane receptors (14). These studies suggested to the present inventors that the volition of clinical resistance to interferon was not always due to changes at the plasma membrane level and might involve altered interaction of nuclear proteins to the

transcriptional regulatory sites 5' to interferon

inducible genes.

A variety of tumors have been shown to be potentially sensitive to treatment with IFN-alpha. These include a range of neoplastic conditions, particularly blood cell related cancers or tumors such as chronic myelogenous leukemia (CML), chronic lymphocytic leukemia (CLL), hairy cell leukemia, sezary syndrome, and mycosis fungoides, as well as lymphomas such as follicular or nodular lymphoma (23). IFN-alpha has also shown promise in the treatment of other cancers such as myeloma (24). Currently, the principal U.S. FDA approved indications for IFN-alpha include CML and hairy cell leukemia.

Unfortunately, it has generally been observed that only a fraction of cancer patients having cancers of a type sensitive to IFN-alpha will in fact exhibit a positive response to IFN-alpha treatment. For example, IFN-alpha therapy results in resolution of most

hematologic abnormalities in about 30% of early CML patients, with half of these exhibiting a complete

disappearance of cytogenetically detectable leukemia cells (10). Yet, overall about 80% of the patients who receive IFN-alpha are ultimately found to have a disease which does not respond to interferon with the development of major cytogenetic remissions. Without cytogenetic

remission, cures are remote. Thus, these 80% or so of cancer patients who receive IFN-alpha therapy, receive it unnecessarily.

An additional complicating factor is the general inability, using current technology, to readily identify those individuals having a tumor that will ultimately respond to IFN-alpha therapy. This is true even in cancers, such as CML, for which IFN-alpha therapy might be indicated. One problem is that in many patients

ultimately resistant to treatment, an initial beneficial effect will often be observed with IFN-alpha therapy. For example, many IFN-alpha resistent CML patients often initially exhibit an improved hematologic profile when they undergo a course of treatment with IFN-alpha. These patients are unnecessarily treated for up to several months with IFN-alpha, making cost a significant factor. There are up to 8000 new cases of CML each year in the United States. If 6800 of these patients are treated unnecessarily, the total cost of the IFN-alpha therapy alone would be almost $100 million for 3 months of

therapy.

A need is therefore readily apparent for an assay having the ability to distinguish those patients bearing IFN-alpha sensitive cancers from those patients that might best benefit from an alternative therapeutic regimen. Not only would such as assay be important economically, but would provide the clinician with information important in deciding upon the most effective course of treatment for that particular individual. The present invention

addresses this shortcoming in the art by providing such an assay.

The present invention embodies in a general and overall sense, the inventors' finding that DNA segments bearing a particular nucleotide sequence have the ability to identify cancer patients having IFN-alpha sensitive cancers. Generally speaking, the invention rests on the inventors discovery that these DNA segments promote the formation of a unique, detectable "complex" with proteins from individuals whose cancers are amenable to treatment with IFN-alpha. This complex, believed to be formed of protein and DNA, is termed a low mobility complex, or LMC, due to its appearance as a "low mobility shift" in a gel mobility shift assay. The LMC is characterized in particular aspects of the present invention as a protein/DNA complex that appears as a low mobility shift away from a faster moving, high mobility complex, or "HMC". One or more such complexes are observable following incubation of the DNA segments with proteins extracted from peripheral blood cells of individuals, depending of whether the proteins are from normal individuals, or from IFN-sensitive or insensitive individuals. Generally speaking, normal individuals (i.e., cancer free) exhibit both an LMC and an HMC.

However, cancer patients, particularly CML patients, tend to exhibit only an HMC, with the LMC disappearing. Upon administering a dose of IFN-alpha to the patient's blood, though, the LMC will reappear in patient's that are sensitive to IFN-alpha treatment. In particular aspects of the invention, the complexes are observed by subjecting incubated admixtures of the extracted protein and DNA to molecular weight

fractionation in order to separate and distinguish the LMC from the HMC. It is believed that virtually any form of fractionation that can successfully separate protein/DNA complexes without disturbing the ionic interactions can be employed, including, for example, non-denaturing gel electrophoresis, gel chromatography, agarose gel,

electrophoresis, and the like can be successfully employed in the practice of the invention. However, in preferred aspects of the invention, non-denaturing polyacrylamide gel electrophoresis is employed, in a technique known as a gel retardation assay. Molecular weight fractionated complexes can be readily observed by a number means, including visualization of the DNA fragment or the bound protein through the use of a label, such as a radioactive, enzyme-conjugated or biotinylated label on the DNA. In preferred aspects, a radioactive label of the DNA fragment is employed. By using a radioactive DNA ligand, the protein/DNA complexes can be readily visualized by autoradiography of the gels following the performance of the gel retardation assay. The principle advantage to the use of a label in the DNA, particularly a radioactive label, is that it avoids the background that would be inherent in any gel fractionation of nuclear proteins.

Numerous methods and radioactive ligands are known in the art for radiolabeling DNA. Generally, any suitable method for labeling DNA can be used in the practice of the invention. A preferred method for labeling the DNA involves end labeling with ³²P.

The inventors have discovered that the LMC response can be readily identified using proteins extracted from cancer patients whose cancers will ultimately prove to be sensitive to IFN-alpha. All that is required in order to elicit the response in such patients is pretreatment of the patient, or their blood, with a 5 x 10⁶ unit dose of IFN-alpha given subcutaneously. In patients whose cancers will not prove amenable to IFN-alpha therapy, one will observe the HMC but not the LMC. Interestingly, normal individuals are observed to be virtually 100% positive for both the HMC and the LMC. The DNA sequences of the invention have an ability to form the foregoing LMCs when admixed and incubated with proteins which include proteins from IFN-alpha responsive cells from the patient under appropriate conditions. DNA segments useful in connection with the invention contain particular sequence elements which have been selected by the present inventors from a consideration of upstream region sequences of genes such as IFN-alpha inducible antiviral, certain other IFN-inducible genes and from experimentation using DNA fragments. Exemplary genes include the 2,5-oligoadenylate synthase gene (2,5-OAS), the GMCSF gene, the zeta and alpha globin genes, the IL-3 gene, a second region from the zeta globin gene

(designated 57/58) B₂ microglobulin, mouse H-2K^b, HLA-DR, and cytomegalovirus. From a consideration of certain regions of these gene sequences, and extensive

experimentation employing variations of the identified regions, the inventors were able to derive a consensus sequence. This consensus sequence along with secondary regions which can lend additional binding strength, provide a useful blueprint for the design of DNA segments for use in the assay of the invention.

In their most basic sense, these consensus DNA segments, useful in the identification of IFN-alpha sensitive cancer patients, include within their sequence what is referred to as an "Octomeric Region" sequence, defined as follows:

-T-G-W-S-S-K-S-A- wherein S = C or G, W = T or A, and K = T or G. This Octomeric Region sequence is believed to be the most basic sequence to which particular IFN-alpha responsive

protein(s) will bind, with the strength of binding

relating, in part, to the particular Octomeric Region Sequence that is employed. The preferred Octomeric sequence, in terms of binding strength or strength of signal, is -T-G-T-S-S-T-C-A-. It is interesting to note that the inventors have found that the Octomeric Region, as well as associated protein binding regions which may be employed, do not have a polarity requirement in terms of on which strand of DNA the sequence is located, or sequence polarity on that one strand. Thus, the foregoing Octomeric Region sequence could be placed within a DNA segment in a "forward" direction (i.e., as 5'-T-G-W-S-S-K-S-A-3') or in a

"reverse" direction (i.e., 3'-T-G-W-S-S-K-S-A-5') and still function for purposes in accordance with the

invention. For this reason, the sequences have been purposefully set forth in the present specification without a 3'-5' or 5'-3' polarity designation.

Furthermore, the use of a "-" on either side of a sequence denotation herein is intended to signify that the denoted sequence is, or can be, part of a larger sequence.

Conversely, where the denoted sequence does not include a "-" on a particular end, it is intended to denote that that sequence is not part of a larger sequence. From a review of the above sequence formula, it will be appreciated that there are a number of possible variations of the Octomeric Region sequence. A

representative number of the combinations included within the Octomeric Region sequence formula that is given have been tested by the inventors and shown to work in the assay. There are, of course, preferred variations that are believed to have a greater binding potential for the IFN-alpha induced protein(s). For example, while the formula indicates that the sixth nucleotide can be either a T or G, the preferred nucleotide at this position is T. Similarly, while the seventh nucleotide can be either a C or G, it is preferably a C. Moreover, while the number eight nucleotide can be either an A or G, it is preferably an A. There does not appear to be any particular

preferences in the four and five positions, and either can be either a C or G. However, it should be noted that in the number 3 position an A can be employed, but an A in this position is believed to provide only a weak binding, particularly if less preferred bases are employed in the number 7 and 8 positions. Thus, for example, if an A is employed in position 3, one may desire to employ an associated binding element, such as ISRE or FSRS region, as discussed below.

Although fragments incorporating only the Octomeric Region can be successfully employed in the practice of the assay, the more preferred segments incorporate one or more associated binding elements. This is because although the binding between the Octomeric Region and the interferon- induced protein(s) is sufficiently tight to obtain a reliable result, the signal obtained is not as strong as the signal obtained with more preferred constructs which also include an associated binding element. One such element is an Interferon Sensitive Response Element region (ISRE), and another is a Friedman-Stark Response Sequence (FSRS) region. Generally speaking, the overall size of the DNA fragment employed in the assay is ultimately limited only by the medium employed to separate the DNA/protein

complexes and the ability to resolve the DNA fragment itself. This is because in performing the assay, one desires to visualize a difference in migration between the DNA present in the LMC complex and that in the HMC

complex, in order to distinguish between the two

complexes. Since in preferred aspects, the complexes are fractionated in a gel retardation assay employing a polyacrylamide gel, one will desire to employ a size fragment that is readily separable and discernible by polyacrylamide gel electrophoresis. Hence, relatively short fragments, on the order of 20 or so basepairs to about 50 or so basepairs, will find the most utility.

However, the inventors have successfully employed fairly long fragments, up to about 200 basepairs or so, prepared by excision of DNA restriction fragments from recombinant vectors and gel purified. Thus, where desired shorter or longer fragments can be employed, so long as the fragments include at least an Octomeric Region sequence as set forth above.

For this reason, it is believed that fragments ranging in size from about 8 up to about 200 basepairs, or even longer if the gel is to be run for extended periods of time, can be employed in the practice of the invention. However, for convenience of preparation by synthetic means, and for improved clarity and detectability of the LMC, one will desire to employ DNA fragments having a shorter length, e.g., on the order of up to 100 or even only 50 basepairs in length, so long as the fragment is at least long enough to contain the Octomeric region

sequence. Preferably, the DNA fragments employed in the practice of the invention will be on the order of about 20 to about 50 basepairs in length. The DNA segments of the invention may be prepared by any of a number of methods known in the art, including, for example, techniques ranging from chemical synthetic means to excision of DNA restriction fragments from vectors which contain sequences in accordance herewith. The inventors have successfully employed both chemically synthetically prepared DNA fragments as well as fragments isolated from recombinant vectors and each appear to function well in the practice of the invention. For most applications, though, one will likely desire to employ DNA segments prepared synthetically through the use of a DNA synthesizing machine such as an Applied BioSystems

synthesizer which has the capability of synthesizing DNA segments having a desired sequence.

Although fragments incorporating only the Octomeric Region can be successfully employed in the practice of the assay, the more preferred segments incorporate one or more associated binding elements. This is because although the binding between the Octomeric Region and the interferon- induced protein(s) is sufficiently tight to obtain a reliable result, the signal obtained is not as strong as the signal obtained with more preferred constructs which also include an associated binding element. One such element is an Interferon Sensitive Response Element region (ISRE), and another is a Friedman-Stark Response Sequence (FSRS) region. A fragment including the ISRE region alone can be successfully employed in assays of the invention. Due to the poor binding that is observed with such a construct, though, this is not recommended. However, when employed along with an Octomeric region sequence, the ISRE region provides a much improved binding and hence more stable and identifiable LMC. This region is referred to as an interferon sensitive response element, ISRE. The ISRE is characterized generally by two stretches of three "T's", usually separated by two (or three) central nucleotides whose identity is not believed to be particularly crucial. Thus, this ISRE is characterized generally by the formula:

-T-T-T-N-(N)-N-T-T-T- wherein N = A, T, G or C. While the inventors have found that the central nucleotides can be any base and still provide a binding capability, in preferred embodiments the central nucleotides are either a G or C residue.

In preparing DNA segments which incorporate both an Octomeric and an ISRE region, one may position the two regions on either side of each other. Furthermore, it is believed that the two elements can be positioned at any point within the DNA segment thus prepared. However, it will generally be desirable to locate the ISRE within about 5 to about 15 nucleotides of one end of the

Octomeric Region, and preferably not closer than 2.

Typically, therefore, one end of the ISRE sequence will generally be positioned within about 15 to about 2 or so nucleotides of one end of the Octomeric region. It is believed that the spacing nucleotides between the two elements can be essentially any sequence. However, for convenience, one may desire to employ the sequence that are found in the representative construct derived from the 2,5 OAS gene as discussed herein.

DNA segments which incorporate only the ISRE can also be employed directly in the assay. However, the binding ability of the ISRE region alone to IFN-alpha induced protein(s) is quite poor, often requiring undesirably long periods of time before complexes can be identified.

Therefore, one will generally desire to employ the ISRE together with an Octomeric Region sequence. An important utility of the ISRE region, though, is the ability to use the ISRE to improve the binding capability of poorly binding Octomeric Region sequences. Thus, for example, if one chooses to employ an Octomeric region which contains a disfavored sequence, such as an A in position 3, and/or another disfavored nucleotide in one of the other

position, the addition of an ISRE region will generally improve binding.

In further embodiments, DNA segments in accordance with the invention can incorporate a third region,

referred to as a Friedman-Stark Response Sequence, or FSRS. As with the ISRE and Octomeric regions, the FSRS can, if desired, be incorporated into DNA segments and employed for the practice of assays in accordance with the invention. Preferably, however, the FSRS will be employed in conneciton with one or more of the foregoing regions, e.g., incorporated at a position essentially adjacent to the ISRE region. The Friedman-Stark Response Sequence, or FSRS, region is defined as having one of the following sequences:

(a) -c-T-C-C-T-C-C-C-T-T-C-T- (from 2,5 OAS gene);

(b) -c-c-C-T-C-T-C-T-C-T- (from B₂ micro- globulin gene);

(c) -T-T-C-N-G-N-A-C-C-T-C-N-G-C-A-G-T-T-T-C- T-C-C-T-C-T-(N)_n-C-T- (wherein n=1-6)

(FSRS consensus)

As with the ISRE, the foregoing FSRS is preferably positioned essentially adjacent to the ISRE, preferably within about 1 to about 400 nucleotides of one end of the ISRE (which can be overlapping). While the FSRS is preferably employed in addition to the ISRE, there is no reason why one could not employ the FSRS alone or directly with the Octomeric Region sequence.

When all three of the foregoing elements are aligned, the most preferred DNA sequence for use in accordance with the diagnostic assays of the present invention is

obtained. An exemplary sequence from the 2,5 OAS gene, which embodies a combination of all three elements, the Octomeric, ISRE and FSRS elements, aligned consecutively in that order, is represented by the following sequence: a) -G-A-G-G-A-G-G-G-A-A-G-A-C-T-C-C-

FSRS

T-T-T-G-C-T-T-T-G-G-T-T-G-T-C-G-T-C-A-G-G-T- ISRE OCTOMER

This sequence corresponds to nucleotides -113 to -76 of the 2,5-OAS gene (4). Sequences from other genes are believed by the inventors to be similarly useful in practicing the assay of the invention, based on a consideration of their sequences. Sequences from genes such as the GMCSF

(granulocyte-macrophage colony stimulating factor; ref. 25), IL3 (interleukin 3); zeta globin (ref. 26-27), alpha globin (ref. 28-29) genes that have been identified include:

(b) -c-A-T-T-T-T-G-T-G-G-T-C-A-C-C-A-T- (-57 to -41, GMCSF);

(C) -G-G-G-G-T-T-G-T-G-G-G-C-A-C-C-T-T- (-110 to -14, IL3); (d) -c-T-C-C-T-T-G-T-C-C-T-C-A-C-T-G-T-

(-174 to -158, 57/58); (e) -C-C-G-G-T-T-G-T-G-G-G-G-G-A-C-A-T- (-202 to -186, 58/56);

(f) -T-T-G-G-T-T-G-T-C-G-T-C-A-G-G-T-T- (-75 to -91, 2,5 OAS);

(g) -T-C-C-T-T-T-C-C-C-A-C-C-G-C-G-A-C-G- (-328 to -311, alpha globin); and

(h) -C-G-T-C-T-T-G-A-C-C-T-G-A-T-G-T-T- (-283 to -267, zeta globin).

All of the foregoing are measured from their

transcription initiation site, except IL3, which is measured from the ATG translation start signal.

As will be appreciated, with perhaps the exception of the alpha globin-derived sequence (g), each of these sequences fit well into the general sequences scheme set forth above for the consensus "Octomeric Region" sequence. The principal exceptions to the consensus sequence occur in the zeta globin sequence (one position), alpha globin (five positions), and zeta globin (one position) sequence.

The assay of the present invention involves a method for identifying cancer patients sensitive to treatment with IFN-alpha. Of course, as used herein, identifying cancer patients sensitive to treatment with IFN-alpha refers to identifying patients whose cancer is, for whatever reason, reasonably amenable to treatment with IFN-alpha. Thus, in IFN-alpha sensitive patients, IFN- alpha therapy will typically result in at least 15% of patients living beyond 5 years, providing they develop a major cytogenetic remission ( reduction in Philadelphia chromosome positive cells, 15%). More often (70%), shrinkage of spleen and normalization of peripheral blood and marrow will occur without an observed reduction in the percent of cells which are Philadelphia chromosome

positive.

The assay of the invention in an overall sense involves generally three steps. The first step is simply identifying a patient who may be in need of IFN-alpha therapy. The patient's blood is then treated with IFN- alpha in a manner which promotes the ability of nuclear proteins of IFN-alpha sensitive individuals to form the Low Mobility Complex. Lastly, cells from the IFN-alpha treated blood are tested for the ability of nuclear proteins from IFN-alpha sensitive cells thereof to form Low Mobility Complex, through the use of a double-stranded DNA segment which includes at least the Octomeric region sequence within its sequence. The ability to form the LMC indicates that the patient is sensitive to IFN-alpha treatment, i.e., that the patient's cancer will likely be amenable to IFN-alpha therapy.

The initial step of the method, identifying patients who may be in need of IFN-alpha therapy, involves simply selecting patients who are suffering from a cancer of a type that is known or has been found to be reasonably amenable or is otherwise sensitive to IFN-alpha treatment. For example, a variety of cancers have been shown to be treatable with IFN-alpha, with varying degrees of success, including chronic myelocytic leukemia (CML), chronic lympholytic leukemia (CLL), follicular and nodular

lymphoma, hairy cell leukemia, myeloma, mycosis fungoides, Sezary syndrome, as well as others.

Most preferably, the assay for IFN-alpha sensitivity will likely find its greatest usefulness in diseases such as CML and hairy cell leukemia, i.e., essentially those cancers where IFN-alpha has found the greatest utility.

In order to elicit the LMC response, one must next treat the patient, or blood cells obtained from the patient, with an effective dose of IFN-alpha. Where the patient is treated directly, a dose of on the order of about 3 to about 5 x 10⁶ units, administered every day will typically be effective to promote the LMC response in IFN-alpha sensitive individuals. However, it is believed that all that is required in terms of IFN-alpha dosage will generally be 3-5 x 10⁶ units daily. In a typical protocol, the patient is administered 3 x 10⁶ IFN-alpha by subcutaneous injection, in order to promote the appearance of an LMC response in sensitive patients.

Administration of IFN-alpha to the patient in vivo can even be avoided altogether where desired, in that the inventors have discovered that similar effect can be achieved by in vitro treatment of blood cells with IFN- alpha. In these embodiments, it has been found that treatment of blood with an amount of IFN-alpha (e.g., 1000 units/cc) to achieve a level roughly commensurate with that which is achieved by parenteral administration.

Thus, in typical embodiments, about 1000 units IFN- alpha/cc is added to about 1 x 10⁶ of abe to form an admixture, and this admixture is allowed to incubate at 20-37ºC for 20 min to 6 hr, preferably about 2 hours. The final step of the overall method involves testing the patient's IFN-alpha treated blood for the ability of proteins thereof to form a Low Mobility Complex with DNA segments as set forth above. Generally speaking, this step of the method will include a step of extracting proteins from the patients blood, generally using

peripheral blood cells which include MNC and more preferably, relatively purified by Ficoll or elutriation. Since the protein/DNA complex which gives rise to the LMC apparently involves the complexion of protein(s) of nuclear origin, one will generally desire to obtain a nuclear protein fraction from the peripheral blood cells. The method for obtaining relatively purified nuclear proteins from peripheral blood cells is not believed to be particularly crucial, so long as there is minimal lysis of nuclei during the initial step. However, the inventors have generally found that the most desirable isolation technique is that taught by Miskimens et al. (15). This technique is preferred because it is less labor intensive. While nuclear proteins isolation is not an absolute requirement, whole cell lysis is messy, and it is believed that the overall results would not be as desirable.

Proteins obtained from the patient are admixed with DNA segments carrying the appropriate binding regions and incubated in a manner effective to form the DNA/protein complexes, such as a salt concentration of 80 mM to 2 M, preferably about 120 mM, DNA concentration ranging from about 1 x 10^-8 g to about 1 x 10⁶ g, preferably about 1 x 10^-6 g, and nuclear protein concentrations of from about 2 to 10, preferably about 4.

An essentially neutral pH is desirable, and will therefore likely find it useful to include a buffer such as Tris or Hepes, preferably Tris. The admixture also preferably includes components such as glycerol and Triton X-100, and is incubated at a temperature of about 4ºC for about 10 to 60, and preferably at least about 10 minutes, in order to promote DNA/protein complex formation.

In particular embodiments, the formation of

DNA/protein complexes, and the identification of LMC formation if one has been formed, is detected through the use of a mobility shift assay. A mobility shift or gel retardation assay, involves subjecting the protein/DNA admixture to electrophoresis through a gel matrix such as polyacrylamide, under conditions wherein any DNA/protein complexes which may have formed, remain stable.

Preferably, the admixture is electrophoresed on a non- denaturing, 5 to 10% polyacrylamide gels.

Materials separated from the admixture, such as by gel electrophoresis, are then visualized, preferably by detecting a radioactive label that has been incorporated into the DNA fragment employed in the assay. When the DNA is radiolabeled, mobility shifts in fragments can be readily detected, e.g., through detecting the appearance of the slower migrating Low Mobility Complex by mens of autoradiography, scintillation counting, or the like.

In still further embodiments, the invention involves the use of the Octomeric region DNA sequence, alone or including one or more of the ISRE or FSRS regions, in the preparation of a double stranded DNA segment, for the purpose of identifying cancer patients sensitive to IFN- alpha. The underlying DNA synthetic techniques are well known in the art and the segments can, in fact, be readily prepared by commercially available DNA synthesizing machines. The selection and preparation of these DNA segments for this particular use is believed to be a novel use of these sequences. Figure 1. Mobility-Shift Gel Electrophoretic

Analysis of Complexes Formed Between the Nuclear Proteins and a DNA Segment Carrying All Three Binding Elements. The DNA Segment employed in the assay had the sequence a) -G-A-G-G-A-G-G-G-A-A-G-A-C-T-C-C-

FSRS

T-T-T-G-C-T-T-T-G-G-T-T-G-T-C-G-T-C-A-G-T-G- ISRE OCTOMER which corresponded to sequences located -113 to -76 nucleotides 5* relative to the translation initiation site of the 2,5-OAS gene. Nuclear proteins were isolated from peripheral blood cells or cell lines by the method of Miskimens et al. (15). The mobility-shift assay was performed as described by Triesman (20), except that the binding buffer contained 120 mM NaCl, 10 mM Tris, pH 7.5, 20 mM 2-mercaptoethanol, 1 mM EDTA, 4% glycerol, and 0.1% Triton X-100. The nuclear protein DNA complexes were stable over a wide range of salt concentrations. The 120 mM NaCl provided a stringent condition for reducing nonspecific interaction. Furthermore, all protein-binding assays were performed after the unfractionated nuclear proteins (2-12 ug) had been incubated with a final

concentration of 1 mg/ml of poly (dl) poly (dC) for 15 min. at 4ºC. The nuclear proteins were then exposed for one hour at 4ºC to 100 fmol or 10⁵cpm of the sequence (a) fragment oligonucleotide ³²P end-labeled by the Klenow reaction, and electrophoresed in a 10% polyacrylamide gel at 5 V/cm for 12 hours. The proteins of the cytoplasmic cellular fraction were compared with the proteins of the nuclear fraction in the gel retardation assay to show that the nuclear protein fractions used in our studies were not contaminated by cytoplasmic proteins. Oligonucleotides were synthesized using an ABS synthesizer, and the sequence of the products verified by Maxim-Gilbert

sequencing reactions.

(A) Competition binding studies of the complexes formed between nuclear proteins and 100 fmol of a ³²P

Klenow end-labeled oligonucleotide. Competition assays were performed by using unlabeled oligonucleotides in a 50-fold (lanes 1, 2, 3, and 5) or 100-fold (lanes 4 and 6) molar excess of the amount of labeled sequence (a)

oligonucleotide to the incubation mixture with HL60 nuclear proteins. To lane 1, an unlabeled oligonucleotide containing the binding site for the SP-1 transcription factor (no homology with any of the binding regions) was added. The SP-1 oligonucleotide did not prevent formation of nuclear protein complexes. To lane 2, unlabeled oligonucleotide from the 5' region of the bacterial alkaline phosphatase gene (no homology with the sequence (a) segment) was added. No inhibition of binding was seen. To lanes 3 and 4, an unlabeled oligonucleotide with the sequence CTGAGACACCAGTCTGAG from the 5' region of the zeta globin gene (similar to the octomer ACAGCAGT present in the interferon-inducible element) was added. This oligonucleotide completely inhibited formation of nuclear protein complexes with the sequence (a) segment. To lanes 5 and 6, an unlabeled oligonucleotide from the 5' region of the zeta globin gene which was not homologous to the interferon-inducible element. This oligonucleotide did not inhibit formation of nuclear protein complexes with the DNA fragment.

(B) Complexes formed between the sequence (a) fragment and the nuclear proteins of normal lymphocytes (lane L), which form a low-mobility complex, and the nuclear proteins of normal monocytes, which form a high- mobility complex. The nuclear proteins of normal myeloid also form a high-mobilitv complex. The fractions of normal lymphocytes or monocytes, which were generated by elutriation of mononuclear cells collected from the peripheral blood of normal donors by continuous-flow centrifugation, were 95% pure as tested by the periodic acid-shift or non-specific esterase stains, respectively.

(C) Gel mobility-shift analysis of complexes formed between the sequence (a) segment and nuclear proteins isolated from the following established human leukemia cell lines: the K562 erythroleukemia cell line (lane 1), the EM2 chronic myelogenous leukemia cell line (lane 2), and the SP-5 T-Cell lymphatic leukemia cell line (lane 3).

(D) Mobility-shift analysis of complexes formed by the DNA segment and the nuclear proteins of normal

peripheral blood cells (lane 1) or the peripheral blood cells of a chronic myelogenous leukemia patient (lanes 2 and 3). A 5-fold increase in the amount of protein was added to lane 3 in order to show that the nuclear proteins from this peripheral blood sample does not form the slowly moving complex.

(E) Gel mobility-shift analysis of complexes formed between the sequence (a) segment and the nuclear proteins of the peripheral blood cells of an interferon-sensitive early chronic phase CML patient before (lane 1) or after (lane 2) in vitro exposure to a 1000 units/cc

concentration of alpha-interferon for 1 hours. (F) A cellular protein lysate from a CML patient

(10% in weight) was added to the nuclear proteins purified from the Daudi cell line which have been shown to form the low-mobility complex with the sequence (a) segment. The binding of the CML lysate (lane 1) and the Daudi cell lysate (lane 2) before and after incubation at 22ºC for 1 hour with the sequence (a) segment (lane 3) is shown after electrophoresis of a mobility-shift gel for 12 hours at 5 V/cmd.

Figure 2. Analysis of Molecular Weights of Nuclear Proteins Which Bind to the Sequence (a) segment in Low- Mobility Complex by Oligoaffinity Column Chromatography and Gel Electrophoresis. The sequence (a) segment

oligonucleotide was covalently bound to biotin by UV irradiation and then bound to a strepavidin agarose column as reported by Triesman (20). The nuclear proteins of a cell line in which the nuclear proteins form the low- mobility complex were bound to the column in 60 mM KCl, 12% glycerol, 10 mM Hepes, pH 7.9, 4 mM Tris, and 1 mM EDTA. After extensive rinsing in this binding buffer, the proteins were eluted at 500 mM KC1, 20 mM Tris, pH 7.0, 5 mM MgCl₂, 1 mM EDTA, and 12% glycerol. The eluted

proteins were run on a 7.5% Lammeli polyacrylamide gel which was developed with silver stain. Figure 3. Photoaffinity Labelling of the Proteins in

Direct Contact With the Seguence (A) Segment in the Low- Mobility complex. The sequence (a) segment was Budr- substituted and ³²P dCTP labeled by a synthetic reaction (Ecoli Polymerase I). The complex formed between the sequence (a) segment and the nuclear proteins of a cell line, which formed only the low-mobility complex, was exposed to UV irradiation and analyzed on a 7.5% Lammeli polyacrylamide gel after reduction and DNase I digestion as outlined by Triesman (20). Similar 84-kD nuclear proteins were detected by this assay from all cell lines tested in which the low-mobility complex formed (HEL, K562, and SP-5) or from interferon-treated CML cells.

Figure 4. DNase I Footprinting Analysis of the

Nucleotide Binding Sites of the low-mobility Complex. The Sequence (A) segment oligonucleotide ³²P 5' end-labeled by the Klenow reaction, was bound to the nuclear proteins of a cell line which formed the low-mobility complex. The oligonucleotide was exposed to 0.005 to 5 ug/mg of DNase I (Worthington, RNase free grade) for 60 sec at 22ºC, following the methods described by Jones et al. (21). The reaction products were phenol chloroform extracted, ethanol precipitated, and analysed on a 20% (19:1

acrylamide: bis) polyacrylamide 8 M urea gel. Lane 1 contains the ³²P 5' end-labeled Sequence (A) segment oligonucleotide before DNase I treatment. Lanes 2 , 3 , 4 , and 5 contain the Sequence (A) segment oligonucleotide treated with DNase I in the absence of nuclear protein at decreasing DNase I concentrations (1/30, 1/60, 1/120, and 1/1200 dilutions of a 0.5 mg/ml solution of Worthington DNase I). Lanes 6-8 contain 20 ul of the nuclear protein incubated with the Sequence (A) segment and subsequently treated with the 1/30, 1/60, and 1/120 dilutions of DNase I, respectively. Regions of protection and

hypersensitivity are marked to the right side of lane 8.

Figure 5. Photoaffinitv Labeling of Nuclear Proteins to the TGTCGTCA or CCTTTCGTTTGG Sites of the 2 , 5-OAS Gene. Nuclear proteins from the cell lines HL60 (lanes 1 and 5), peripheral blood cells of a CML patient (lanes 2 and 6), K562 (lanes 3 and 7) and Daudi (lanes 4 and 8) were incubated with a synthetic oligonucleotide which contained the TGTCGTCA (lanes 1-4) or CCTTTCGTTTGG (lanes 5-8) labeled with ³²P and substituted with Budr synthetically. Following exposure to ultraviolet light, the protein DNA complexes were electrophoresed on 7.5% NaDodSO₄ PAGE gels in 10% beta mercaptoethanol after boiling for 15 minutes. The nuclear proteins seen in lane 1-4 is 84 kDa.

The present invention evolves out of the inventors' studies on the DNA binding of nuclear proteins, and in particular, out of studies which appeared to show that the binding of nuclear proteins to a particular upstream region of interferon inducible genes is different when nuclear proteins from CML cancer patients are employed than they are when proteins from normal, non-cancerous cells are employed. Additional studies undertaken in light of these findings indicated to the inventors that IFN-alpha "corrected" the binding pattern of nuclear protein binding in IFN-sensitive CML patients, but did not correct the binding pattern in non-IFN-sensitive cancer patients. From these findings the inventors postulated that an assay could be developed which could readily and reliably identify patients who could benefit from IFN- alpha therapy by identifying those whose cancer would by responsive to IFN-alpha therapy. The present invention is the culmination of the inventors undertaking in this regard, and it can now be demonstrated that a sensitive and dependable assay has been developed.

As discussed above in the Summary, the assay in a general sense involves simply testing peripheral blood cells from a cancer patient who has received an initial dose of IFN-alpha for the ability of nuclear proteins from these cells to form a Low Mobility Complex (an "LMC") with a DNA segment bearing particular sequence characteristics. The ability of nuclear proteins from peripheral blood cells to form such an LMC is diagnostic in that under the conditions of the assay, normal individuals, as well as cancer patients responsive to IFN-alpha, will form such a LMC whereas non-responsive individuals will not. The presence or absence of an LMC is determined through the use of a mobility shift assay, wherein extracted proteins are bound to a selected DNA fragment and then separated by gel electrophoresis. The appropriate sequence characteristics and

variables of the DNA segments employed in the assay have been extensively studied by the inventors, who have identified a particular eight-basepair "consensus" sequence which appears to be the basic binding DNA

sequence for binding IFN-responsive protein(s). This basic eight basepair sequence is referred to by the inventors as the "Octomeric Region" sequence, and its sequence characteristics, including all of the possible sequence variations, have been reviewed above in the summary.

Additional sequence regions have also been identified which can be employed in conjunction with the Octomeric Region in preparing DNA segments for use in the assay. These regions, the so-called Interferon Sensitive Response Element and the Friedman-Stark Response Sequence, can be incorporated in DNA segments along with the Octomeric Region in order to improve the DNA binding function of the segment. As mentioned, the most preferred DNA segments for the purposes of carrying out the assay will generally be segments which contain all three regions, such as represented by DNA segments which include nucleotides - 117 to -73 nucleotides 5' relative to the translation initiation site of the 2,5 OAS gene (see sequence (a) above).

The ISRE and FSRS are believed to be contact points for nuclear proteins, a binding that is altered by IFN- alpha. Presumably, these receptors are inducing the transcriptional activation of a gene, but the identity of this gene is unknown. In a practical sense, the assay may be performed generally as follows:

1. interferon Alpha Treatment

to Elicit IFN Response

After identifying a candidate patient, which will generally be a patient suffering from a cancer of a type amenable to IFN-alpha therapy, the patient's blood is treated with a dose of IFN-alpha that is effective to elicit the LMC response, if such a response is

forthcoming. In order to treat the blood cells with IFN, IFN-alpha may be administered directly to the patient, or, alternatively, blood cells from the patient may be treated with IFN-alpha in vitro. In either case, all that is required is a short, initial course of IFN-alpha

treatment.

Where one elects to administer IFN-alpha directly to the patient in order to elicit the response, it will generally be desirable to administer a normal treatment dose of IFN-alpha to the patient subcutaneously, with the subcutaneous route being preferred, for 1 to 2, preferably 2, days. As will be appreciated by those of skill in the art, a typical IFN-alpha treatment dose range, and a dose range that will useful in the practice of this aspect of the invention, will be on the order of 3 to 10 x 10⁶ units, with about 5 x 10⁶ units s.c. being preferred.

After the appropriate treatment of the patient with IFN, about 20 to 40 ml of blood sample is removed for use in the next step of the assay.

Where one elects to treat blood cells with IFN

in vitro to elicit the response, the treatment time is somewhat shortened. In this embodiment, a 20 to 40 ml blood sample is removed from the patient. The blood sample is then heparinized to prevent coagulation and the blood is further prepared by separating the WBC from RBC using a commercially available mononuclear cell

fractionation media (e.g., commercial Ficoll). An

appropriate dose of IFN-alpha is then admixed directly with the heparinized blood cells and this admixture allowed to incubate for 2 to 3 hours, preferably at about 20 to about 37 degrees C. Useful IFN-alpha doses for in vitro treatment will generally range from about 1000 units/cc to about 2000 units/cc, with about 1000 being preferred.

These IFN-alpha treated blood cells, whether treated in vivo or in vitro, may then be employed in the next step of the assay, which requires isolation of peripheral blood lymphocytes, followed by extraction of nuclear proteins from the isolated lymphocytes.

2. Nuclear Protein Extraction from

Peripheral Blood Lymphocytes

After blood cells have been obtained from the patient and IFN-alpha treated as described above, one will next generally desire to fractionate the blood to obtain relatively purified mononuclear cells (MNC). The problem with using whole blood which includes RBCs in the assay is essentially twofold: the use of unfractionated blood makes hemoglobin the major protein component of the admixture, which makes it difficult to ensure that a proper amount of nuclear protein is present in the

incubation mixture, and the hemoglobin tends to make the polyacrylamide gel difficult to read. The obtaining of relatively purified nuclear proteins from MNC is further important in order to avoid or reduce the possibility of non-specific binding or cross binding of non-specific proteins to the DNA fragment employed in the assay. In order to isolate the peripheral blood MNC from the IFN-treated blood, one preferably first subjects the blood to fractionation by passing the cells over a ficoll gradient into white cells, serum and red cells. The white cells are the desired MNC fraction. The use of Ficoll gradients for the isolation of white blood cells is generally well known in the art and there are a number of procedures and commercially available gradients which can be employed. In a preferred technique (discussed more fully in the Examples below), a MNC separation medium from Organon Teknicka Corp. is employed. While Ficoll gradient fractionation is a preferred means for MNC preparation, other techniques are known in the art and can be employed, and even continuous flow centrifugation and elutriation (ref. 30).

The relatively purified MNC are then employed for the isolation of nuclear proteins. This involves first lysing the cells followed by isolation of nuclei, and then the preparation of proteins from the isolated nuclei. Numerous techniques are known in the art for nuclear protein isolation, and the inventors see no reason why any such technique would not be successful in the practice of this aspect of the invention. The inventors have found, though, that the nuclear protein isolation technique as described by Miskimens et al. (Ref. 15) is particularly preferred because the technique is not labor intensive and the total protein yield is high. An exemplary procedure that is routinely employed by the inventors to prepare peripheral blood cell nuclear proteins from blood is set forth below in Example I.

In general, the preferred technique involves first lysing the isolated cells in buffer solution containing a nonionic detergent such as Triton X-100 and Sucrose. This lysing buffer is employed to lyse the cell membrane without lysing the nuclear membrane, thus allowing the removal of cytoplasmic proteins prior to lysing the nuclei. This cell lysate containing unlysed nuclei is then subjected to centrifugation in order to pellet the nuclei, following which the nuclear pellet is resuspended in the lysing buffer. Additional agents such as

spermidine and a relatively high salt (e.g., about 1.5- 0.75 M) are added and the mixture incubated on ice in order to deproteinate and precipitate the DNA. This nuclear lysate is then subjected to ultracentrifugation in order to pellet the cellular debris, DNA, RNA and

insoluble materials, leaving the nuclear proteins in solution. The supernatent from this high speed spin is then retained and dialyzed into an appropriate buffer.

After the nuclear proteins have been prepared, dialysed, etc., they are ready to be employed in the nuclear protein binding assay as discussed below.

3. Mobility Shift Assay

The binding assay involves simply admixing the isolated nuclear proteins with a radiolabeled DNA segment which includes an appropriate DNA binding sequence, followed by fractionation of the incubated material on a non-denaturing polyacrylamide gel. There are, however, some additional procedures which serve to reduce nonspecific background and thereby improve the quality of the results. In general, one will desire to first admix a small amount of the nuclear protein, for example, on the order of about 4 to 8 micrograms of protein, with an amount of poly dl/poly dC (Pharmacia) (or poly dl'dC), and incubate this admixture on ice for several minutes (15 minutes is adequate). Incubation of the nuclear proteins with poly dl/poly dC is a useful step because the poly dl/poly dC serves to reduce subsequent non-specific binding of proteins to the labeled DNA segments. Poly dl/dC binds proteins which bind double stranded DNA in a sequence independent manner.

The incubation of protein with DNA is preferably performed at a relatively high salt concentration, on the order of 80 mM to 2 M, with about 120 mM salt being preferred. Although not strictly required, the binding assay is preferably performed at such salt concentrations because, as with the poly dl/poly dC preincubation, the use of a relatively high salt concentration tends to reduce non-specific interactions and bindings. However, it is specifically pointed out that the inventors have found that the LMC DNA/protein complex is stable over a wide range of salt concentrations (e.g., 80 mM to 2 M). Thus, it is believed that the salt concentration employed is not crucial and, if desired, one can employ a salt concentration in the range of about 50 mM to about 120 mM.

After preincubation of the nuclear proteins with poly dl/poly dC in binding buffer, an amount of an appropriate radiolabeled DNA is added. Since one will preferably be employing a radioactive label in the DNA, and the

DNA/protein complexes will be subsequently visualized by means of this DNA label, the actual amount of DNA employed is not believed to be particularly important. In fact, it will generally be advantageous to employ only enough DNA to provide an appropriate detectable signal on the

subsequent polyacrylamide gel. For example, the present inventors routinely employ ³²P end-labeled DNA and have found that only about 1 to 5 x 10⁵ cpm of ³²P counts

(which generally corresponds to on the order of about 100 fmol, or about 1 x 10^-8 g, of DNA) are entirely adequate. One should be careful not to overload the reaction with

DNA as this will tend to obscure the results. Therefore, an upper limit of about 1 x 10^-6 g of DNA should generally be observed.

The DNA/protein admixture is then incubated,

preferably on ice, at least initially, and otherwise maintained at about 4 to 20 or so degrees C, for about 30 minutes to 1 hour in order to allow the DNA/protein interactions to form and come to equilibrium. After incubation, the material is subjected to gel

electrophoresis on a non-denaturing polyacrylamide gel by whatever technique is convenient. A preferred gel electrophoresis protocol for the practice of the mobility shift assay employs a 5, 7.5 or 10% polyacrylamide gel (preferably 10%) electrophoresed for about 12 to 48 hours in a selected non-denaturing buffer such as a 10 mM

EDTA/33 mM NaOAc/67 mM Tris pH 7.5 buffer, until the BPB marker dye has migrated to about 20 cm on the gel. To visualize the gel migrated protein/DNA complexes, the gel is then autoradiographed for an appropriate period of time, usually on the order of about 2 hr to about 16 hr.

Alternatively, one may find it advantageous to employ 0.25 TE buffer and 150 volts for 4 hours. This is faster and fits into a single work day, but the gel is of poorer quality.

4. Identification of the Low Mobility Complex

After the gel has been autoradiographed so as to visualize the gel-migrated radioactive DNA, the LMC, if present, can be readily identified by visual inspection. It is not absolutely necessary, although it is preferred, to run a control (such as a known negative or positive sample, or a sample obtained from the patient prior to IFN-alpha induction) in that each sample will have an

"internal" control in the form of the HMC (High Mobility Complex). Thus, positive samples will exhibit a

characteristic "doublet" of bands upon autoradiography, with the faster migrating band being the HMC, and the slower moving band the LMC. Although the apparent size of each of the bands will vary according to the size of the DNA segment that is employed for the binding assay, as well as the gel and electrophoresis conditions, it has generally been found that when a DNA fragment of about 45 basepairs is employed and the assay performed under the conditions set forth in Example II below, the LMC will migrate at about one-fourth the rate of the free DNA, and the HMC will migrate at about a rate which is 20% faster than the CMC. 5. Exemplary Experimentation

The following examples are included to provide actual working protocols which the inventors routinely employ in carrying out aspects of the invention. For example, Examples I and II set forth standard protocols that are employed for performing the nuclear protein extraction and mobility shift assay, respectively. In addition to

Examples I and II, Example III is included to demonstrate the application of these techniques in both an

investigative and analytical context. Those of skill in the art should readily appreciate that many of the techniques employed in the following Examples are

illustrative of standard laboratory practices found by the inventors to work well in the practice of the invention. It will be apparent to those of skill in the art in light of the following Examples that numerous changes in materials and/or modifications in procedures can be made and nevertheless achieve a useful result. EXAMPLE I NUCLEAR PROTEIN EXTRACTION a. Materials

The materials and buffers listed below are those normally employed in the inventors' laboratory for

preparation of peripheral blood cells and for carrying out the nuclear protein extraction. The catalogue numbers of preferred reagents have been included for the convenience of the reader.

Beckman SW 50.1 or Sorvall AH-650 rotor cooled to 4ºC Ultracentrifuge cooled to 4ºC

LSM-lymphocyte separation medium

*I903*Misc. 50 and 15 ml P.P. Falcon tubes

Several 1.5 ml microcentrifuge tubes

2-P-1000 pipetman

1-P-200 pipetman

1-P-20 pipetman

Vortex genie

Microcentrifuge

Low speed centrifuge

Buffer II

Buffer II is employed to lyse the peripheral blood cells once they have been separated from RBCs, and further it is employed to lyse the nuclei cells prepared from the lysed cells by centrifugation.

40 ml 100 mM Hepes pH 8.0 (Fisher-BP310-500)

4 ml 5 M NaCl (Sigma-S-3014)

200 ul 0.5 M EGTA pH 8.0 (Sigma-E-4378)

50 ul 0.5 M EDTA pH 8.0 (Sigma-E-5134) 2 ml 1 M MgCl2 (Fisher-BP214-500)

2 ml Triton X-100 (Sigma-T-6878)

After the foregoing reagents are admixed, the

admixture is brought to 360 ml with sterile, preferably microfiltered, water, autoclaved and stored at 4ºC until needed. Before using, the following additional reagents should be added to the buffer admixture: 0.171 g sucrose (Sigma-S-0389)

1 Ul of 1 M DTT (BRL-5508UA)

0.1 ml 100 mM PMSF (Sigma-P-7626)

1 ul 5 mg/ml aprotinin (Sigma-A-1153) This buffer can now be referred to as Buffer II complete, and should be used within 30 minutes or made up fresh again.

Buffer III

Buffer III is a buffer that is employed for dialyzing the extracted nuclear proteins prior to employing the proteins in the DNA binding/gel mobility shift assay. The buffer is prepared by admixing the following reagents:

50 ml of 100 mM Hepes pH 8.0

0.5 ml of 1 M MgCl2

50 ml of 5 M NaCl

225 ml of sterile water

125 ml sterile glycerol (warmed before pouring about a min. in the microwave) (Sigma-G-7757)

Before using Buffer III, one should add 1 ul of 1 M DTT/1 ml and 1 ul of 100 mM PMSF/1 ml. b. Procedures

The following procedures are those generally employed in carrying out the peripheral blood cell selection and protein extraction. Please note that all procedures should be carried out on ice or at 4ºC.

1. First, about 7 ml of LSM (Organon Teknika

Corporation, cat. #36427/841001) is placed into a 50 ml tube. Then, about 20 ml of peripheral blood containing 20 unit/ml of heparin is layered over the LSM. The LSM-layered blood is then spun at about 4000 rpm in a 4ºC in Technospin type analytical centrifuge for 1 hour.

2. After spinning, both the buffy coat, which appears at the top of the gradient, as well as the middle layer of the gradient, are slowly drawn off. 3. The material which has been recovered from the

gradient is placed into a 50 or 15 ml tube (depending on the volume of the recovered material) and again spun at 4000 rpm at 4ºC, for about 10 minute in the Technospin like centrifuge in order to pellet the cells.

4. After pelleting the cells, the supernatant is poured off, and the pellet resuspended in 1 ml of PBS

(Fisher, cat #MT-21-031-lv) which contains 20 units of heparin/ml.

5. The resuspended material is then placed into a

microcentrifuge tube, such as a 1.5 ml tube. Each tube or sample can then be processed individually. 6. To process each tube, the following steps are

performed: a. The tube(s) is spun in microfuge for 15 seconds in order to pellet the cells. b. The supernatant is then drawn off and the pellet resuspended in 0.5 ml of Buffer II (complete) with vigorous mixing, followed by vortexing for about 15 seconds. c. The vortexed admixture is then respun in the

microfuge for an additional 15 seconds. d. The supernatant is then drawn off the pellet

quickly before the pellet has an opportunity leak nuclei, and the pellet resuspended in 0.5 ml of Buffer II (complete). e. About 6 ul of 1 M Spermadine (Sigma S-2501) is then added to the resuspension and it is mixed gently, after which 75 ul of 5 M NaCl is added and the suspension mixed further. f. The suspension is then allowed to sit on ice for about 60 minutes, and then transferred to appropriate ultracentrifuge tubes (e.g., Beckman 326819 Sorvall 03127). 7. The centrifuge tubes containing the nuclear

suspension are then centrifuged at 30,000 rpm for 60 minutes in a Sorvall AH-650, in order to pellet cellular debris, DNA, RNA, etc. 8. In order to prepare to dialyze the extracted nuclear proteins after the spin is finished, one will want to cut dialysis tubing into approximately 6 inch pieces, and rinse the tubing with deionized water. Of course, one will want to ensure that the tubing does not contain holes. The tubing is then placed into Buffer III to equilibrate it prior to dialysis.

9. After the spin has finished, the supernatant is

poured into the dialysis tubing, which is then tied and tagged, e.g., with colored plastic clips.

10. The supernatant is then dialyzed against Buffer III (complete) with a change after 24 hours.

11. Samples are then stored in a convenient container, such as a 1.5 ml microcentrifuge tube, aliquoting about 0.5 ml/tube. About 10 ul of 50 mM Benzamidine (Sigma-B-6505) and 3.4 ul of 2-mercaptoethanol

(Sigma-M-6250) are added to each 1.5 ml tube, and the material stored at -20"C pending use.

STOCK SOLUTIONS

The following are recipes for various of the stock solutions employed in the foregoing procedures.

100 mM Hepes 500 pH 8.0

11.9 gm pH to 8.0 with NaOH

qs to 500 ml

autoclave

5M NaCl 1 liter

292 gm

qs to 1 liter

autoclave

0.5 M EDTA 100 ml PH 8.0

19.02 gm in 60 ml millipore water

add 3-4 pellets of solid NaOH

pH to 8.0 with NaOH

qs to 100 ml

autoclave

0.5 M EGTA 100 ml PH 8.0

18.6 1 gm in 60 ml millipore water

add 3-4 pellets of solid NaOH

pH to 8.0

qs to 100 ml

autoclave

1 M MgCl2 100 ml

20.33 gm

qs to 100 ml millipore water

1 M DTT 10 ml

15 g

qs to 10 ml

filter sterilize aliquote and freeze

100 mM PMSF 50 ml

0.87 1 gm in 50 ml isopropanol

store at -20 (stable in this form)

1 M spermadine 5 ml

1.27 gm into 5 ml millipore water EXAMPLE II

MOBILITY SHIFT ASSAY

a) Materials

The following materials and buffers are those commonly employed in the inventors' laboratories for performing the mobility shift assay and are included herein in the following manner for the convenience of the readers.

The following Sample Buffer is employed for binding DNA segments with extracted proteins, and can be made up in advance (except for the 2-mercaptoethanol addition) and stored until use.

10 X Sample Buffer

1 % Triton X-100 1 ml

40% glycerol 40 ml

10 mM EDTA 2 ml, 0.5 M

100 mM Tris, Ph 7.5 10 ml, 1 M

1 M NaCl 20 ml, 5 M After mixing the above ingredients, the volume is brought to 100 ml with water and autoclaved. Just prior to use, one should add to the buffer about 14 ul of fresh 200 mM 2-mercaptoethanol per milliliter of buffer. The following electrophoresis buffer is the preferred buffer for performing the polyacrylamide gel

electrophoresis of the DNA/protein complexes and can be prepared in advance.

10 X Electrophoresis Buffer

10 mM EDTA 20 ml, 0.5 M

33 mM NaOAc (anhydrous) 2.7 gm

67 mM Tris pH 7.5 67 ml, 1 M

After the reagents are admixed, the mixture is brought to 1 liter with water, and autoclaved (note the best results occur when the final pH is 7.7-7.9). Other reagents used in the assay:

1 % Bromσphenol Blue

Dissolve 100 mg of BPB in 10 ml of water

and sterile filter.

10 ug/ul poly dl * poly dC

see instructions from supplier (Pharamacia)

b) Procedures

The following are the inventors ' standard procedures for carrying out the mobility shift assay. 1. First, in order to allow the gel to be pre- electrophoresing while the binding assay is being conducted, one will generally desire to prepare preferably a 5, 7.5 or 10% acrylamide gel, and preferably with a cross linker ratio of about 10 g acrylamide to 0.2 g bis. The gel is prepared using the 10 x electrophoresis buffer above.

2. After the gel is poured, it is pre-electrophoresed at about 3 mA per gel (20 X 40 gels at 80 volts; 20 X 20 gels at 40 volts).

3. It is generally convenient to number the tubes that will be employed (preferably microfuge tubes) and the tubes allowed to cool in ice bucket.

4. About 1 ul of 10 X sample buffer is added to each

tube, followed by the addition of 1 ul of 1 X poly dI * poly dC. 5. The volume in each tube is adjusted with sterile

water so that when the protein is added in the next step, the final volume will be 10 ul. 6. Next, about 4-8 ug of nuclear protein is added to the tube, bringing the final volume to about 10 ul, and the admixture incubated on ice for 15 minutes. 8. Then, about 1- 500,000 cpm of labeled double stranded DNA segment is added to the incubated mixture, and the mixture allowed to incubate an additional 30 minutes to 1 hour on ice. 9. One ul of 1% Bromphenol blue is then added to each sample, and the samples are loaded into well of the gel. The gel is electrophoresed for 12-48 hrs at 3 mA per gel. After electrophoresis is complete (BPB to 20 to 40 cm), the gels are autoradiographed for about 2 hr to 10 hr.

EXAMPLE III

Interferon-Induced Binding of an 84-kD Nuclear Protein to a Specific DNA sequence as a Marker for IFN-alpha Sensitivity in CML

The present example sets forth a series of studies which have been undertaken using the most preferred DNA segment for forming the DNA/protein complexes, the

*I903*upstream region, -133 to -79 5' of the initiation site, of the 2, 5-OAS gene. (It has also been shown that the 6-16, 9-27 and beta-2-microglobulin genes also behave in a manner which is identical to the transcriptional enhancers of the 2,5 OAS gene). These studies are

exemplary of the practice of an embodiment of the assay.

A gel mobility-shift assay such as is set forth in Example II above, was employed to study the formation of complexes between nuclear proteins and DNA segments incorporating sequences present 5' to the 2,5 OAS gene.

These sequences included: (i) the Friedman-Stark Response Sequence, which is found in enhancer elements 5' to Class I and HLA antigen genes; (ii) the interferon response element or ISRE which is found in transcriptional enhancer 5' to interferon-inducible antiviral genes, 2,5- oligoadenylate synthetase gene (2,5-OAS), and (iii) a third region which the present inventors have determined important to binding capability, the Octomeric Region sequence. The 2,5 OAS contains elements similar to elements from other interferon-inducible genes (ISG15,

ISG54, IP10, G-16, and G-27) the function of which is as yet unknown (16).

In order to test the sequence-specificity of the binding of nuclear proteins to the 2,5 OAS derived DNA segment, (corresponding to the sequence (a), nuclear proteins from various sources were incubated with ³²P Klenow end-labeled DNA in the presence or absence of unlabeled oligonucleotide sequences. The incubated material was then tested for the formation of nuclear protein DNA complexes by the mobility-shift assay. As shown in Fig. 1A, addition of a 50-fold molar excess of an unlabeled segment (a) (lanes 5-6, Fig. 1A), completely inhibited binding of nuclear proteins to the ³²P-labeled fragment. In contrast, when a 50-fold molar excess of unlabeled non homologous oligonucleotides were used (lanes 1-4, Fig. 1A), no reduction of binding or nuclear protein to the ³²P-labeled segment occurred. These experiments show that the binding of nuclear proteins to the DNA segment corresponding to segment (a) is sequence-specific under the conditions of the assay.

When the DNA segment was incubated with nuclear proteins isolated from peripheral blood lymphocytes of normal individuals or from lymphoid cell lines, a single slowly moving nuclear protein DNA complex formed (see Figs. 1B-C). This complex is referred to as the low- mobility complex. Nuclear proteins isolated from normal monocytes or myeloid cells or from myeloid-monocytoid cell lines formed a more rapidly migrating complex with the segment (see Figs. 1B-C). This complex is referred to as the high-mobility complex.

The nuclear proteins of normal peripheral blood cells (20% lymphoid cells and 70% myeloid cells) formed both the low-mobility and the high-mobility complexes with the segment in 17 of the 17 normal individuals tested (see

Table 1 and 2 and Fig. 1D). The nuclear proteins of 47/53 of untreated CML patients formed only the high-mobility

•

complex and not the low-mobility complex (see Fig. 1D), and the low-mobility complex which appeared in the 4 CML patients positive for this complex was very faint. Studies were also conducted to determine the in vivo effect of alpha-interferon on the binding of nuclear proteins from CML patients who were sensitive or resistant to alpha-interferon. Following alpha-interferon therapy (5 X 10⁶ units/day), the peripheral blood of 57/58 of interferon-sensitive CML patients contained nuclear proteins which formed the low-mobility as well as the high-mobilitv complexes with the sequence (a) segment (see Table 1) whereas, only the high-mobility complex formed with nuclear proteins from cells of CML patients not receiving interferon. The changes induced by alpha- interferon reversed 48 hours after removal of alpha- interferon (see Table 2).

No increase in the percentage of lymphoid cells was seen in many of the patients in whom in vivo exposure to alpha-interferon led to the appearance of the low-mobility (lymphoid) nuclear protein complex with the sequence (a) segment (see patients #1, 2, 3, 4, and 8 of Table 2B). Therefore, any differences seen would not be attributable to changes in the numbers of lymphoid, erythroid,

monocytoid, and myeloid elements in the marrow which result from IFN treatment. Interferon induction of the low-mobility complex occurred within 72 hours of in vivo exposure in some patients (#4, 6, and 41 of Table 2).

None of the samples of nuclear proteins isolated from the peripheral blood of the 17 resistant CML patients formed the low-mobility complex with the sequence (a) segment after interferon exposure. The nuclear proteins of these interferon-treated-resistant patients formed only the high-mobility nuclear protein complex. Peripheral blood cells were collected from an

interferon-sensitive patient during a short interruption of interferon therapy, and were treated in vitro with 1000 u/cc of alpha-interferon for one hour. This resulted in a change in the binding of nuclear proteins form that shown in Figure ID for untreated CML patients (high-mobility complex) to that shown in Figure ID for normal peripheral blood cells (low- and high-mobility complexes). The low-mobility complex was not detectable in the peripheral blood of untreated CML patients even in the presence of 20% lymphocytes (see patient #2, Table 2B), whereas, the low-mobility complex was detectable on the peripheral blood of CML patients with only 1% lymphocytes after exposure to alpha-interferon. Percoll fractions of CML marrow, in which lymphocytes were enriched and myeloid cells were decreased, contained the low-mobility complex. This suggested that the mixture of Philadelphia positive myeloid cells and Philadelphia negative lymphoid cells which co-exist in CML peripheral blood led to loss of the low-mobility complex.

To test this directly, a protein lysate from CML peripheral blood was added (10% in weight) to nuclear proteins purified from a lymphoid cell line which had been shown to form the low-mobility complex, and incubated these extracts in the presence of protease inhibitors for 60 minutes at 22ºC. As shown in Figure 1E, the intensity of the low-mobility complex decreased and new high- mobility complexes appeared in the presence of the CML protein lysate. This effect was not seen when the incubation was set at 4ºC or when it was conducted for only 5 minutes at 22ºC instead of for 60 minutes. To determine the molecular weights of the nuclear proteins associated with the alpha-interferon-inducible low-mobility complex, the nuclear proteins of a cell line in which only the low-mobility complex is formed with the segment was fractionation on an oligoaffinity column which contained the sequence (a) segment as a binding ligand.

The proteins which bound to this column were

electrophoresed into a 0.1% SDS, 7.5% Lammeli

polyacrylamide gel under reducing conditions (7). As shown in Figure 2, the molecular weights of the nuclear proteins which bind to the segment in the low-mobility complex were 73- and 84-kD. Photoaffinity-labeling studies (see Fig. 3) suggested that the 84 but not the 73- kD protein was in direct contact with the binding elements present in the segment, as only the 84-kD nuclear protein became ³²P-labeled by this procedure.

It was then determined whether nuclear proteins extracted from a cell line that forms only the low- mobility complex, protected the interferon regulatory element from DNase I digestion. As shown in Figure 4, these nuclear proteins protected 2 regions of the segment from digestion by DNase I: one of which contained the ISRE sequences GAAACG, and another which contained the 3' end of the Friedman-Stark Response Sequence CCTCCCTTC.

These 2 protected regions were separated by a short stretch of hypersensitivity, TGAG.

As shown in Figure 3, DNAse I footprinting analysis indicated that binding of the low-mobility complex

produced both regions of protection and hypersensitivity, the latter perhaps due to a distortion or bending of the double-stranded DNA in the region of the short R loops. At the center of this region is the Octomeric sequence TGTCGTCA (See Figure 3), a variation of which has been identified in the 5' regulatory regions of a number of hematopoietic genes (globin, GM-CSF, IL3) as well as the transcriptional enhances of SV40, polymer and adenovirus. This region is bounded by short, direct repeats (GGT) located 11 nucleotides inside a short inverted repeat in a region that has been shown to be important in the

transcriptional activity of the 2,5-OAS gene. Immediately 5' to this TGTCGTCA region is a region CCTTTGCTTTGG, which is protected from DNAse I digestion by nuclear proteins in the binding assays (See Figure 3) . This region is similar to a region found to be important in the induction of transcription by interferon (2,5).

In order to assess the binding of nuclear proteins to the CCTTTGCTTTGG and TGTCGTCA domains of the 2,5-OAS regulatory region, the binding of nuclear proteins of hematopoietic cells to synthetic oligomers which contained only the CCTTTGCTTTGG 5' interferon response element, or the TGTCGTCA sequence was studies. As shown in Figure 5, the cross-linking and the gel retardation assays of nuclear proteins to the TGTCGTCA sequence were

indistinguishable from that seen with the entire 2,5-OAS regulatory sequence, whereas very little binding occurred to an oligomer which contained the CCTTTGCTTTGG domain alone. Since the CCTTTGCTTTGG domain is protected in the footprinting assay (Figure 3), this indicates that the binding of nuclear proteins to the two sites (CCTTTGCTTTGG and TGTCGTCA) is cooperative.

Thus, IFN-alpha induced a change in the binding of nuclear proteins following in vivo exposure of the peripheral blood cells in 57/58 interferon-sensitive CML patients. None of the 17 interferon-resistant CML patients studied exhibited changes in the binding of nuclear proteins to the DNA segment after exposure to alpha-interferon. The interferon-induced changes in nuclear protein complex formation which were observed may be due to its effect of an activity present in untreated CML cells which alters binding of nuclear protein with the binding regions. A correlation between in vitro

sensitivity to alpha-interferon-induced changes in the binding of nuclear proteins has recently been reported in interferon-sensitive and resistant B-cell lines (16).

IFN-alpha induced changes in nuclear protein binding to the transcription regulatory regions of interferon- inducible genes may be playing a role in mediating the clinical effects of IFN-alpha.

The in vitro effect of IFN-alpha on nuclear protein binding occurs within 1 hour and is not associated with changes in the percentages of lymphoid or myeloid cells in the peripheral blood of CML patients. Studies by other workers have shown that transcriptional activation by IFN- alpha takes place within 5 minutes of its binding to the plasma membrane interferon receptor (17). Furthermore, the interferon-induced binding of nuclear proteins to the transcription regulatory regions of these genes is

insensitive to cycloheximide inhibition (3) and is

correlated temporarily with the activation of gene

expression (18).

The portion of the Friedman-Stark consensus complex and the ISRE which were protected in the DNase I assays in the studies shown in Fig. 4 are present in a number of other interferon-inducible genes including 6-16 which has been reported to be inducible in interferon-sensitive but not resistant cell lines (19). If the clinical effect of interferon involves transcriptional activation of a gene through induction of binding of these nuclear proteins to interferon transcription regulatory elements similar to the ones which have been studied, it is possible that several interferon-inducible genes may be involved.

Studies are underway the present inventors to identify which interferon-inducible genes are important in

mediating the clinical effects of interferon in CML.

This nuclear protein binding assay is clearly useful in predicting the responsiveness of CML patients to alpha- interferon therapy. The assay is unlikely to predict sensitivity to gamma-interferon as the recognition

elements involved in mediating the effect of gamma and alpha-interferons are different. Further studies are focusing on the characterization of the 84-kD protein in both normal individuals and CML patients, which may serve as a more direct target for the assay.

* * * * *

The present invention has been disclosed in terms of preferred modes found to work well in the practice of the invention. However, numerous modifications and changes in the steps, procedures used and materials will become apparent to those of skill in the art in light of the disclosure. For example, changes in the manner in which the LMC DNA/protein complex will become apparent.

Moreover, there is no requirement that the DNA segment employed in the binding assay be labeled radioactively. Similar, modifications and improvement in the techniques used for isolation and binding proteins from the patients will be apparent in light of the disclosure. These and other modifications are intended to be within the spirit of the present invention and scope of the appended claims.

* * * * * REFERENCES

1. Joklik, W.K. (1986), "Interferons", Chapter 15 in Fundamental Virology, eds. Fields et al., pp. 281- 307.

2. Sugita et al. (1987), Mol. Cell Biol. , 7:2625-2630.

3. Benech et al. (1987), Mol. Cell Biol. , 7:4498-4504.

4. Rutherford et al. (1988), EMBO J. , 7:751-759.

5. Reich et al. (1988), Proc. Natl. Acad. Sci. USA,

84:6394-6399.

6. Newell et al. (1960), Science, 132:1497-1498.

7. Kurzrock et al. (1988), N. Enol. J. Med., 319:990- 998.

8. Groffen et al. (1984), Cell, 36:93-99.

9. Konopka et al. (1984), Cell, 37:1035-1042. 10. Talpaz et al. (1986), N. Engl. J. Med., 314:1065- 1069.

11. Zullo et al. (1985), Cell, 43:783-800. 12. Berreck et al. (1985), The Embo J., 4:2249-2256.

13. Triesman et al. (1987), The Embo J., 6:2711-2717.

14. Maxwell et al. (1985), Int. J. Cancer, 36:23-28. 15. Miskimens et al. (1985), Proc. Natl. Acad. Sci. , USA, 88: 6741-6744

16. Kessler et al. (1988), EMBO J., 7:3279.

17. Friedman et al. (1984), Cell, 38:745.

18. Hannigan et al. (1986), EMBO J., 5:1607. 19. McMahon et al. (1986), J. Virol., 57:362.

20. Triesman (1986), Cell, 46:567.

21. Jones et al. (1987), Cell, 48:79.

22. Jaira et al. (1985), J. Interferon Res., 5:583-596.

23. Strauder, H, (1986), Adv. Cancer Res., 46:1-265. 24. See Cooper, M.R., Seminars in Oncology, 15:21-25, for examples.

25. Miyatake et al. (1985), EMBO Jrnl., 4:2561-2568. 26. Laver et al. (1980), Cell, 20:119-130.

27. Proudfoot et al. (1982), Cell, 31:553-563.

28. Chang et al. (1977), Proc. Natl. Acad. Sci. USA,

74:5145-5149.

29. Liebhaber et al. (1989), Proc. Natl. Acad. Sci. USA,

22:7054-7058. 30. Turpin et al. (1988), J. Clin. Apheresis, 3:111.

Claims

CLAIMS:

1. A double stranded segment of DNA of from about 8 to about 200 basepairs in length, the segment capable of forming a low mobility complex with nuclear proteins of peripheral blood cells from alpha interferon-sensitive individuals, the DNA segment including within its sequence an Octomeric region sequence as follows: -T-G-W-S-S-K-S-A- wherein S = C or G, W = T or A, and K - T or G.

2. The DNA segment of claim 1, wherein the Octomeric region sequence comprises:

-T-G-T-S-S-T-S-A-

3. The DNA segment of claim 2, wherein the Octomeric region sequence comprises:

-T-G-T-S-S-T-C-A-

4. The DNA segment of claim 2, wherein the Octomeric region sequence comprises: -B-B-T-T-G-T-S-S-T-S-A- wherein B = G, C or T.

5. The DNA segment of claim 1, wherein the Octomeric region comprises:

-T-T-G-T-S-S-K-S-A-

6. The DNA segment of claim 1, wherein the Octomeric region sequence is selected from the group of sequences consisting of: a) -T-T-G-T-C-C-T-C-A-;

b) -T-T-G-T-G-G-G-C-A-;

c) -T-T-G-T-G-G-T-C-A-;

d) -T-T-G-T-C-G-T-C-A-; and

e) -T-T-G-A-C-C-T-G-A-.

7. The DNA segment of claim 1, wherein the segment includes within its sequence a sequence selected from the group of sequences consisting of: a) -A-A-T-G-A-A-T-A-A-G-G-A-C-G-G-T-

G-C-A-G-A-A-C-T-G-G-A-C-T-A-C-A-N-; b) -T-G-T-A-A-G-G-C-C-A-C-A-G-G-A-G-A-G-G-A- A-C-A-G-G-A-G-T-G-A-C-A-G-C-C-C-C-C-A-A-;

C) -T-C-C-C-C-C-C-G-C-C-T-T-G-C-C-C-

G-G-G-G-T-T-G-T-G-G-G-C-A-C-C-T-; and d) -T-T-c-C-C-C-C-G-C-C-T-T-C-C-C-T-G- G-C-A-T-T-T-T-G-T-G-C-T-C-A-C-C-A-.

8. The DNA segment of claim 1, which, in addition to the Octomer region, comprises an Interferon Sensitivity

Response Element (ISRE) region as follows: -T-T-T-N-(N)-N-T-T-T- wherein N = A, T, G or C.

9. The DNA segment of claim 8, wherein the ISRE region comprises:

-T-T-T-S-S-T-T-T-

10. The DNA segment of claim 8, which, in addition to the Octomer region and the ISRE region, comprises a Friedman- Stark Response Sequence selected from the following: -C-T-C-C-T-C-C-C-T-T-C-T-;

-C-C-C-C-T-C-T-C-T-; and

-T-T-C-N-G-N-A-C-C-T-C-N-G-C-A-G-T-T-T-C-T-C- C-T-C-T-(N)_n-C-T- (wherein n=1-6)

11. The DNA segment of claim 10, wherein the segment is characterized as including the following sequence:

-G-A-G-G-A-G-G-G-A-A-G-A-C-T-C-C-T-T-T-G- C-T- T-T-G-G-T-T-G-T-C-G-T-C-A-G-T-G-

12. The DNA segment of claim 1, defined as being a fragment of from about 8 to about 100 basepairs in length.

13. The DNA segment of claim 1, defined as being a fragment of from about 8 to about 50 basepairs in length.

14. The DNA segment of claim 13, defined as being a fragment of from about 20 to about 50 basepairs in length.

15. A double stranded DNA segment, of from about 8 to about 200 in length, the segment including a sequence selected from the group of sequences consisting of:

(a) -G-A-G-G-A-G-G-G-A-A-G-A-C-T-C-C-T-T-T- G-C-T-T-T-G-G-T-T-G-T-C-G-T-C-A-G-T-G-;

(b) -C-A-T-T-T-T-G-T-G-G-T-C-A-C-C-A-A-;

(c) -G-G-G-G-T-T-G-T-G-G-G-C-A-C-C-T-T-;

(d) -C-T-C-C-T-T-G-T-C-C-T-C-A-C-T-G-T-;

(e) -C-C-G-G-T-T-G-T-G-G-G-G-G-A-C-A-T-;

(f) -T-T-G-G-T-T-G-T-C-G-T-C-A-G-G-T-T-;

(g) -T-C-C-T-T-T-C-C-C-A-C-C-G-C-G-A-C-G-;

(h) -C-G-T-C-T-T-G-A-C-C-T-G-A-T-G-T-T-;

(i) -C-T-C-C-T-C-C-C-T-T-C-T-;

(j) -C-C-C-C-T-C-T-C-T-; and

(k) -T-T-C-N-G-N-A-C-C-T-C-N-G-C-A-G-T-T-T-

C-T-C-C-T-C-T-(N)_n-C-T- (wherein

N=A,T,C,G, and n=1-6).

16. The DNA segment of claim 15, wherein the selected sequence comprises sequence (a).

17. The DNA segment of claim 15, wherein the selected sequence comprises sequence (b).

18. The DNA segment of claim 15, wherein the selected sequence comprises sequence (c).

19. The DNA segment of claim 15, wherein the selected sequence comprises sequence (d).

20. The DNA segment of claim 15, wherein the selected sequence comprises sequence (e).

21. The DNA segment of claim 15, wherein the selected sequence comprises sequence (f).

22. The DNA segment of claim 15, wherein the selected sequence comprises sequence (g).

23. The DNA segment of claim 15, wherein the selected sequence comprises sequence (h).

24. A method for identifying cancer patients sensitive to treatment with alpha-interferon, the method comprising the steps of:

(a) identifying a patient in need of alpha- interferon therapy;

(b) treating blood cells of the patient with alpha- interferon; and

(c) testing the alpha-interferon treated blood cells for the ability of proteins thereof to form a Low Mobility Complex, through the use of a double- stranded DNA segment which includes within its sequence an Octomeric region sequence as follows:

-T-G-W-S-S-K-S-A- wherein S = C or G, W = T or A, and K = T or G, the ability to form such a complex indicating that the patient is sensitive to treatment with alpha interferon.

25. The method of claim 24, wherein the patient has chronic myelogenous leukemia, hairy cell leukemia, chronic lymphocytic leukemia, myeloma, mycosis fungoides, Sezary syndrome, follicular lymphoma or nodular lymphoma.

26. The method of claim 25, wherein the patient has chronic myelogenous leukemia.

27. The method of claim 24, wherein step (b) comprises the removal of a blood sample from the patient and treatment of the removed blood sample with alpha- interferon, after which the alpha-interferon-treated sample is tested for the ability of proteins therein to form the complex.

28. The method of claim 24, wherein step (b) comprises the administration of a effective amount of alpha- interferon to the patient, after which a blood sample subsequently removed from the patient is tested for the ability of nuclear proteins therein to form the complex.

29. The method of claim 24, wherein the patient's blood is tested by a method which includes the steps of:

(a) extracting nuclear proteins from peripheral blood cells from the patient's blood;

(b) incubating the proteins with the DNA segment under conditions effective to promote selective protein/DNA binding; and

(c) testing for the formation of a Low Mobility Complex.

30. The method of claim 29, wherein step (c) comprises subjecting the protein-incubated DNA to a gel mobility shift assay.

31. The use of an Octomeric region DNA sequence as follows:

-T-G-W-S-S-K-S-A- wherein S = C or G, W = T

or A, and K = T or G, in the preparation of a double stranded DNA segment, the DNA segment for identifying cancer patients sensitive to treatment with alpha-interferon.

32. The use of claim 31, further comprising the use of an Interferon Sensitive Regulatory Element region DNA

sequence as follows:

-T-T-T-N-N-T-T-T- wherein N = A, T, G or C, together with the Octomeric region sequence in the preparation of the DNA seqment.