Acute promyelocytic leukemia

OBJECTIVES: To identify the chromosomal translocation common in M3 and discuss its diagnostic use to:

Compare acute leukemia with chronic leukemia and other forms of cancer.

Describe the molecular defect including the fusion gene and fusion protein produced from the translocation.

Discuss the proposed mechanism of leukemogenesis in M3.

Discuss the proposed mechanism of differentiation induction stimulated by ATRA therapy.

Present the future direction of this and other forms of therapy. DATA SOURCES: Current literature.

DATA SYNTHESIS: Acute promyelocytic leukemia (AML-M3) is a form of acute leukemia that presents with a less dramatic leukorytosis, anemia, and thrombocytopenia than other acute leukemias. However, AML-M3 has a lower first remission rate and a higher morbidity and mortality rate than most of the other acute leukemias when treated with conventional chemotherapy. AML-M3 frequently stimulates a serious concomitant coagulation disorder, disseminated intravascular coagulation, which is a major contributor to the high mortality rate. This and other devastating sequela of M3 have prompted clinicians and investigators to develop methods of improving diagnosis and therapy. In 1977 the method of diagnosis confirmation was improved by the identification of a consistent chromosomal translocation involving the long arms of chromosomes 15 and 17. Identification of the specific molecular lesion that produced the t(15;17) translocation occurred in 1990 and was shown to involve the retinoic acid receptor alpha gene (RARalpha). Because the RARalpha gene is mutated in all AML-M3 patients studied so far

and because it is often the only mutation identified, several proposed mechanisms of leukemogenesis have evolved. From these discoveries a novel approach to cancer treatment focusing on differentiation therapy instead of traditional chemotherapy emerged. Alltrans retinoic acid (ATRA) has been shown to stimulate differentiation of promyelocytes from the malignant clone and has become an important element in the treatment of patients with AML-M3.

CONCLUSION: Since the discovery of the t(15;17) transloca

tion, the identification of the fusion gene containing the retinoic acid receptor or" and the success of ATRA as a form of differentiation therapy, the diagnosis and treatment of AML-M3 has dramatically improved. In addition, AML-M3 has become a model system used to study the mechanisms that produce uncontrolled growth and lack of differentiation in leukemic cells (leukemogenesis) and the mechanisms of therapeutic reversal of this block in differentiation (differentiation therapy).

ABBREVIATIONS: (N-CoR, Sin3, HDAC) = retinoic acid tran

scription co-repressor complex; AML = acute myelocytic leukemia; APL = acute promyelocytic leukemia; ATRA = all-trans retinoic acid; cDNA = complementary deoxyribonucleic acid; DIC = disseminated intravascular coagulation; DNA = deoxyribonucleic acid; Eco RI = Escherischia coli restriction endonuclease 1; FAB = French-American-British; FISH = fluorescence in situ hybridization; M3 = acute promyelocytic leukemia; mRNA = messenger ribonucleic acid; oligo dt = oligomer deoxythymidine; PML = promyelocytic leukemia; POD = PML Oncogenic Domain; poly A = polyadenosine; RA = retinoic acid; RARa = retinoic acid receptor alpha; RNA = ribonucleic acid; rRNA = ribosomal ribonucleic acid; rtPCR = reverse transcriptase polymerase chain reaction; RXR = another retinoid receptor; tRNA = transfer ribonucleic acid; UV = ultraviolet WBC = white blood cell

INDEX TERMS: acute myelocytic leukemia; acute promyelocytic leukemia; AML-M3; all-trans retinoic acid (ATRA); PML gene; PML/RARalpha fusion gene; RARalpha gene; t(15;17)

Clin Lab Sci 2000;13(2):106

Focus Continuing Education Credit: see pages 123 to 126 for learning objectives, test questions, and application form.

LEARNING OBJECTIVES

1. Draw and describe the chromosomal translocation commonly observed in AML-M3.

2. List two other chromosomal defects observed in AML-M3 patients.

3. Describe the molecular defect in AML-M3 to include the genes and chromosomes involved.

4. Draw the exon map and discuss the normal function of the RARa gene.

5. Discuss the normal structure and proposed function of the PML gene.

6. Describe the structure and draw the exon map of the PML/ RAROC fusion gene and the fusion protein product.

7. Discuss the proposed mechanism of leukemogenesis induced by the PML/RAROC fusion protein.

8. Discuss three commonly used molecular methods of detecting the PML/RARa fusion gene and state the one most useful to detect minimal residual disease.

9. Discuss the most promising proposed mechanism of ATRA induced differentiation of promyelocytes in patients with AML-M3.

10. Compare the differences in response to ATRA verses 9-cisRA based on the breakpoint in the PML gene.

GLOSSARY

Allele - Any one of two or more genes that occupy the same locus on a chromosome.

Avidin - A protein from raw egg white that is used in biological assays for its ability to specifically bind biotin.

Biotin - The D isomer component of the vitamin Bz complex required by most organisms and used in biological assays for its ability to specifically bind avidin.

Centromeres - The nonstaining primary constriction of a chromosome that separates the short arm from the long arm. Complementary - When two strands of nucleic acids, DNA or RNA, possess their bases in an appropriate sequence that allows them to bind due to base pairing.

Conserved sequences - DNA sequences that are common across species that represent normal, wild-type sequences.

DNA probe - A short nucleic acid sequence that is labeled with radioactivity or fluorescence and used to identify a gene bearing a complementary base sequence.

Eukaryotic - Higher complexity organisms including protozoans, fungi, plants, and animals that are characterized by membrane bound nucleus, chromosomes, and mitochondria and divide by mitosis.

Exons - The coding sequences within a gene that are transcribed into mRNA and translated into protein.

Flanking sequence - A DNA sequence that abuts the sequence of interest.

Heterodimer - Two different molecules that preferentially bind one another and usually perform a common function.

Homodimer - Two identical molecules that preferentially bind one another and usually perform a common function.

Hybridize - The selective binding of two nucleic acid strands based on their complementary base sequences.

Intron - The non-coding sequences of a gene that are transcribed into RNA, spliced out of the mRNA, and are not represented in the protein product.

Isoforms - Molecules that possess the same function but exhibit minor differences in chemical composition.

Ligand - A molecule that specifically binds to a macromolecule, often a receptor, and initiates a function that is usually observable in the cell bearing the receptor or measurable in a fluid.

Novel - A formerly undiscovered object (in molecular biology a gene) with an unknown function.

Oncogenes - Any of a family of genes that under normal circumstances code for proteins involved in cell growth or regulation but may foster malignant processes if mutated.

Promoter - A DNA sequence at which RNA polymerase binds and initiates transcription.

Restriction endonuclease - One of many enzymes isolated from bacteria that cleaves double stranded DNA chains at interior bonds and at specific sequences.

RXR - A DNA binding protein in the retinoic acid receptor family that normally dimerizes with RARa, binds RA response elements on DNA, and stimulates transcription.

Somatic - All non-reproductive cells of the body.

Stringency - The controlled environment (salt concentration, formamide concentration, and temperature) in which DNA hybridization experiments are conducted. A high-stringency hybridization reaction requires perfect base pair complementarity between the probe and the target gene. In contrast, hybridization can occur with several base pair mismatches in a low-stringency environment.

Synergistic - Coordinated or correlated action of two or more agents so that the combined action is greater than the sum of each acting separately.

Transfected - Cells that have received a foreign gene from a carrier vector, usually a retrovirus, that results in integration and expression of the gene.

Transform - The conversion of a cell from normal to malignant resulting from one or more genetic mutations.

Tumor Suppressor Gene - A normal gene that functions to control growth that, when mutated, results in unregulated growth.

Ubiquitous - Expressing a widespread presence in nature or in a particular species.

Wild-type - The sequence of a gene as it naturally appears in the species without mutation.

INTRODUCTION

Acute promyelocytic leukemia (APL) is classified as an acute myelocytic leukemia subclass M3 (AML-M3), using the FrenchAmerican-British (FAB) classification system.1,2 AML-M3 represents a malignant transformation of a hematopoietic progenitor cell that results in clonal proliferation and lack of differentiation of promyelocytes in bone marrow and peripheral blood. The disease is one of the more serious versions of acute leukemia expressing a rapid onset of symptoms, poor response to chemotherapy, a frequent association with disseminated intravascular coagulation, and a high morbidity and mortality rate.1,2 In 1977, a translocation involving chromosomes 15 and 17 was identified and has become an important diagnostic indicator for AML-M3.3-5 The genes involved in this t(15;17) translocation were identified in 1990 as the retinoic acid receptor alpha (RARalpha) gene on chromosome 17 and the promyelocytic leukemia gene (PML) on chromosome 15.6-9 This genetic lesion produces a chimeric fusion gene that is transcribed into a stable mRNA and translated into a fusion protein.10-12 This review will discuss the t(15;17) translocation and its use diagnostically; describe the molecular lesion that produces the uncontrolled proliferation and loss of differentiation; describe a proposed model of leukemogenesis; and discuss a proposed mechanism of drug induced differentiation using all-traps retinoic acid (ATRA).

CHROMOSOMAL DEFECT IN AML-M3

AML-M3 is characterized by a consistent, balanced, reciprocal translocation involving the long arms of chromosomes 15 and 17, referred to as t(15;17)(Figure 1).3-5 In a balanced reciprocal translocation, a single break in the DNA occurs in two different chromosomes resulting in an equal and opposite exchange of genetic material between the two chromosomes. If the chromosome pieces are large enough, usually greater than 1,000,000 base pairs, the altered chromosome morphology can be appreciated microscopitally. The t( 15;17) translocation is identifiable by karyotype analysis in greater than 70% to 80% of M3 cases, is usually the only chromosomal abnormality observed and has never been demonstrated in any other type of neoplasia.4,5,13-16 There have been a few cases of M3 that possess a translocation involving chromosomes 5 or 11 instead of 15. However, the consistent Feature in these cases is that the partner chromosome thus far, has always been 17.17

As can be seen in Figure 1, chromosome 15 breaks near the midpoint of the long arm at band 22 while chromosome 17 breaks in the long arm at band 21. The two long arm fragments switch chromosomal positions producing a lengthened chromosome 15 and a shortened chromosome 17.3-5,11,16-17

As stated earlier, this translocation can be identified by karyotype analysis in greater than 70% to 80% of M3 patients, making the incidence quite high. 13,16 However, in some patients with M3, the translocation is not resolvable at the chromosomal level but is at the molecular level. When combining the chromosomal and molecular methods of identifying the t(15;17) translocation, the mutation is found in nearly all cases of M3.13-16 Occasionally two breaks can occur in a single chromosome or in both chromosomes, one at the same site as before and the other at a different, but proximal, site. These chromosome pieces contain the same genes involved in the original translocation and they can translocate to the other chromosome producing the same fusion gene. However, the quantity of DNA translocated is far too small to be visualized by karyotype analysis. This situation results in a patient with clinical M3 that appears, by karyotype analysis, not to possess the t(15;17) translocation when in fact the mutation is present, and can be identified at the molecular level.

MOLECULAR DEFECT IN AML-M3

Whereas the identification of the t(15;17) translocation is important in the diagnosis of the disease, the molecular product or products of the translocation are critical to the understanding of the leukemogenic process in M3. Because the t(15;17) translocation is found in nearly every case of M3, is not found in any other leukemia, and is often the only mutation identified, suggests that the translocation itself may be the cause of the malignant transformation. Therefore, the genetic material at or near the breakpoint of one or both chromosomes must represent a coding sequence formerly involved in the control of myeloid growth and/or differentiation. It has been determined that the breakpoint on chromosome 15 occurs within a novel gene referred to as PML for promyelocytic leukemia (Figure 2). Chromosome 17 breaks within the gene that codes for a receptor that is sensitive to retinoic acid known as RAROC. This translocation produces the fusion protein PML/RARa.6-9 If the functions of the normal genes are known, then the function of the newly formed fusion protein can be predicted. PML/RARa is successfully transcribed into a stable mRNA molecule that is translated into a fusion protein. In order to predict the transforming function of the fusion gene, we must understand the functions of the normal genes.

NORMAL RARalpha GENE

As stated earlier, the normal RARalpha gene is located on band 21 of the long arm of chromosome 17. The middle panel of Figure 2 illustrates the RARalpha, gene as having six short exons with relatively large introns, especially intron 2 located between exons 2 and 3. Intron 2 is the site of genetic mutagenesis in the RARalpha, gene. The six exons code for six, distinct functional domains present on the RARalpha protein. The first two domains function to modulate transcription in a ligand dependent manner. The third domain binds DNA while the fourth domain may be involved in nuclear localization of the RARalpha protein. The fifth domain is the site of ligand binding. This suggests that when the specific ligand, which is retinoic acid, binds RARa (fifth domain), it stimulates the protein to bind hormone response elements (third domain) on promoter regions of DNA and activate gene transcription (first and second domains). Since disruption of the RARalpha gene results in a block in myeloid differentiation, it can be presumed that the protein products of the genes activated by RARalpha are involved in the differentiation of myeloid cells.5

As illustrated in Figure 3, RARalpha proteins normally homodimerize with other RARalpha proteins or form heterodimers with RXR proteins and each dimer binds to retinoic acid (RA) response elements in the promoter regions of various genes.18 This binding, in the presence of retinoic acid, presumably stimulates the transcription of genes, with the resultant translation of proteins that may be involved in myeloid differentiation pathways. The RARalpha/RXR heterodimer has been shown to be far more responsive to this type of transcription activation than the RARalpha homodimer.19

Recent experiments have shown that in the absence of retinoic acid, RARalpha/RARalpha homodimers or RARalpha/RXR heterodimers bind a large ubiquitous nuclear protein (N-CoR) which causes transcriptional repression when bound by two other nuclear proteins, mSin3 and histone deacetylase (HDAC).19 When retinoic acid binds the dimers, they are released from the repressor molecules and free to bind a transcription activator complex involving histone acetylase (Figure 3). This newly formed activation complex will bind RA response elements on DNA and stimulate the transcription of proteins necessary for myeloid differentiation. It is presumed that normal myeloid differentiation signals will trigger the release of retinoic acid, which will in turn stimulate the release of the repressor complex from the dimer in favor of the activator complex.20-22 This represents a ligand driven differentiation pathway that, when blocked, would result in a break in myeloid differentiation and the subsequent accumulation of malignant promyelocytes.

NORMAL PML GENE

Less is known about the PML gene and the function of the protein produced by this gene. The PML gene is located on band 22 of the long arm of chromosome 15. As illustrated in the upper panel of Figure 2, the gene contains nine coding exons, some of which possess the characteristics of DNA binding domains. Upon transcription, the mRNA undergoes alternative splicing resulting in 16 isoforms of the PML protein that diverge in the carboxy-terminal end.23 The function of the PML protein is not known but the structure would suggest possible roles in the regulation of gene expression, DNA recombination, and DNA repair. Recent evidence suggests that PML is highly expressed in late G1 or S phases of the cell cycle and that it inhibits cellular growth. Therefore, the lack of PML function could result in a loss of growth suppression, thus stimulating cell proliferation.24-28 In normal human bone marrow, PML expression appears to be primarily restricted to myeloid cells and is localized to the nucleus of these cells.5 PML protein has been shown to be a surface component of a large, multisubunit, subnuclear structure called PML Oncogenic Domain (POD). There are at least five different proteins in each POD structure, including PML, and 10 to 20 POD structures per nucleus. POD structures are not found at sites of DNA replication, RNA splicing, centromeres, or nucleoli.2930 It is thought that PODs may be involved in cell cycle control and/ or DNA damage control.22

FUSION GENE

The breakpoints that define the PML/RARa fusion gene can be specifically described. As Figure 2 illustrates, the breakpoint on chromosome 17 occurs in the 5' region of the RARalpha gene in the second intron between exons 2 and 3. Numerous breakpoints have been identified within this breakpoint region of chromosome 17 but because all occur in the third intron, the effect on the fusion gene product is identical. The entire gene sequence is six exons in Length. The loss of the first two exons from chromosomal breakage results in the retention of the last four of the six total exons. The breakpoint on chromosome 15 occurs in the 3' region of the PML gene. There are three breakpoints (bcrl, bcr2, bcr3) that have been clearly defined and are labeled as 3, 2, and 1 in the upper panel of Figure 2. The most common breakpoint (bcrl) occurs between exons six and seven and is labeled as number 1 on the corresponding figure. The other two breakpoints, bcr2 and bcr3, occur in exon 6 and intron 3, respectively. Therefore, the fusion gene represents a head-to-tail joining of the PML gene on chromosome 15 to the RARalpha gene on chromosome 17. The resulting PMh/RARalpha fusion gene contains most of the exons of both genes and is under the transcriptional control of the PML promoter.5-9 As illustrated in Figure 4, the gene is transcribed into a stable mRNA molecule that is translated into a fusion protein.

FUSION PROTEIN

The fusion protein possesses the amino acid sequence that corresponds to the first 6 exons of the PML gene, fused with the amino acid sequence transcribed from the last four exons of the RARalpha gene (Figure 4). As stated earlier, at least three splice sites exist in the PML gene, but two forms dominate. Both isoforms, however, produce proteins that contain the DNA binding domain (cystein rich regions) and the dimerization domain (leucine zipper structures). Since there is a single breakpoint region in the RARalpha gene, the tail of the protein has only one form. This form contains both the DNA binding domain (labeled as DNA on lower panel of Figure 4) and the ligand-binding domain (labeled as RA on lower panel of Figure 4) of RARalpha.6-9

An understanding of the binding and distribution characteristics of the fusion protein is necessary before discussing the proposed mechanism of fusion protein function. PML/RARalpha protein has been shown to localize primarily in the nucleus, and to a lesser extent in the cytoplasm, of transfected cells.31-36 This represents a different localization pattern than either wild-type PML or RARalpha. Although both PML and RARalpha are nuclear proteins, the pattern and area of the nucleus in which the fusion protein localizes is different than both wild-type counterparts. The 10 to 20 POD structures that contain normal PML proteins are disrupted in M3 blasts and promyelocytes and replaced by greater than 100 small, fine, speckled structures that contain PML and PML/RARalpha protein 30,35 In addition, the presence of the PML/RARalpha in the cytoplasm clearly indicates an alteration in cellular distribution, which would obviously affect PML and RARa function. When treated with retinoic acid, M3 blasts, promyelocytes, and equivalent cell lines are stimulated to dramatically reorganize PML containing proteins back to the normal wildtype POD structures.30,32-36 Therefore, fusion proteins that bind other DNA binding sites or establish a home in the cytoplasm, presumably find their way back to the natural PML specific DNA binding sites when exposed to ATRA.

The most promising mechanism of fusion protein leukemogenesis is illustrated in Figure 5. Recent evidence suggests that the PML/ RARalpha fusion protein binds the normal N-CoR, Sin3, HDAC corepressor complex with a higher affinity that does wild-type RARalpha. Therefore, physiological levels of retinoic acid are insufficient to trigger their dissociation resulting in continuous repression of the RARalpha target genes and a block in myeloid differentiation.20-22

Another characteristic of the fusion protein, that affects transcription activation, is its ability to dimerize. As stated earlier, the fusion protein produced by patients with AML-M3 forms homodimers or heterodimers with wild-type PML proteins and with normal RXR proteins, all of which bind RA response elements.32-37 The availability of these proteins to form dimers is directly related to their production by the corresponding genes. Since the transformed promyelocytes are somatic cells, they have two copies of each chromosome. Therefore, each leukemic cell would have one copy of the fusion gene, one copy of the reciprocal fusion gene, one copy of the wild-type PML, and one copy of the wild-type RARalpha. In addition, the leukemic cells would possess two copies of the wild-type RXR gene since RXR is not involved in the t(15;17) translocation. This sets up a competition for protein binding. The fusion protein may come in contact with another copy of the fusion protein, with a copy of the normal RARalpha protein, with a copy of the normal PML protein, or with a copy of the normal RXR protein. These dimer interactions further complicate the role of the fusion protein in AML-M3 leukemogenesis by diverting wild-type PML, RARa and RXR from their desired DNA binding sites. It has been shown that when M3 cells are treated with retinoic acid, RXR returns to a normal cellular distribution while the fusion protein reverts to a pattern typical for wild-type RARalpha.34 This redistribution presumably dissociates the fusion protein from the RA response elements and from RXR. Consequently, both RA response elements and RXR would be freed from the fusion protein allowing each of them to bind RARalpha thus contributing to the restoration of normal function. In summary, not only does the continued binding of the fusion protein to the co-repressor complex block transcription, but the inability of wild-type RARalpha, PML, and RXR proteins to bind RA response elements occupied by the fusion protein prevents normal transcription activation.

MOLECULAR DETECTION METHODS

There are three commonly used methods of detecting the PML/ RARa translocation at the molecular level; fluorescence in situ hybridization, known as FISH: southern blotting; and reverse transcriptase polymerase chain reaction, abbreviated rtPCR. The FISH technique can be performed in at least two different ways: an indirectly labeled, single probe technique and a directly labeled two probe technique. In the indirectly labeled single probe technique, a biotin-labeled DNA probe is commercially available that contains a sequence complementary to the PML/RARalpha fusion gene (upper panel of Figure 6). Blasts and promyelocytes from the M3 patient are collected and air-dried or fined to a microscope slide with methanol and acetic acid. Double stranded DNA is denatured with 70% formamide to produce single stranded material onto which the probe can bind. The probe is added to the cells on the slide in 50% to 65% formamide and incubated. The biotin-labeled probe will hybridize to the PML/RARalpha fusion gene, if present. Avidin, labeled with a fluorescent molecule, is added to the cells on the slide. Avidin will bind the biotin attached to the probe and emit fluorescence.

Following incubation, the fluorescent signal can be amplified by anti-avidin antibodies followed by anti-globulin, tagged with a fluorescent label (amplification procedure not pictured). Cells expressing the PML/RARalpha fusion gene will bind the probe and emit fluorescence while the normal wild-type alleles that lack the fusion sequence will not bind the probe. The color emitted is dependent on the particular fluorochrome label used.38,39

As can be seen in the lower panel of Figure 6, two or more probes, directly labeled with different fluorochromes, can be used concomitantly to identify multiple mutations, or mutations and normal alleles in the same cell. In this case two different probes were used, one that would hybridize with the PML gene found on both the normal PML allele and on the 5' end of the fusion gene and one that is complementary to the RARalpha gene that is also found on both the normal RARa, allele and on the 3' end of the fusion gene. Therefore, each probe would hybridize to their normal alleles while both probes would hybridize to opposite ends of the fusion gene. In the example in Figure 7, the red signal in the upper left quadrant of the nucleus is the normal PML allele on chromosome 15 while the green signal to the far right is the normal RARalpha allele on chromosome 17. The combination red/green signal, located between the other two signals, is the fusion gene.38,39 The two-probe technique has the advantage of requiring only one set of two probes since it binds conserved sequences while the one probe technique requires a different probe for each mutation.

The FISH technique has a major advantage over karyotyping analysis because a small percentage of patients with AML-M3 possess a form of the mutation that is resolvable at the molecular level, but karyotype negatively. The FISH technique would reduce the number of false negatives that would result from the use of karyotype analysis alone. In addition, FISH is generally considered less difficult to perform than either rtPCR or southern blot and it is comparatively rapid. However, FISH is limited to the identification of only particular, known mutations, but is reasonable specific and sensitive.

Southern blot remains the gold standard to detect this mutation and most other genetic defects at the molecular level. In this technique, illustrated in Figure 8, leukemic cells are collected from blood or bone marrow and resuspended in buffer. In order to release the DNA, the cells are lysed with the detergent sodium dodecyl sulfate (SDS). The DNA is extracted from the cellular debris into the upper aqueous layer using a phenol:chloroform solution. The DNA in the supernatant is precipitated with ethanol and resuspended in buffer. The DNA is digested with restriction endonucleases to reduce the genomic DNA from long continuous segments, into small manageable fragments. These fragments are then electrophoresed in a denaturing agarose gel with a DNA control, to separate them based on size. The DNA fragments are transferred or blotted to a nylon or nitrocellulose paper and baked to bind the single stranded DNA to the paper. A single stranded, sequence complementary DNA probe, labeled with radioactive molecules or enzymes is mixed with a buffer, added to the blot and incubated overnight. The probe will selectively hybridize to complementary sequences, if present, amongst the DNA fragments. In this case, the sequence being detected would be the conserved sequences of either the PML or RARalpha gene. Slots are washed in salt and SDS to remove unbound probe and to establish a binding environment to control the specificity of the reaction, referred to as stringency. High-stringency washes using low salt concentrations and high temperatures, prevents binding of DNA with base mismatches unlike low-stringency washes. Hybridization can be visualized by measuring color development on the blot when enzyme labels are used or by developing the radioactive blots onto autoradiographs. Figure 9 illustrates an autoradiograph from an M3 patient. The left gel represents the DNA of an M3 patient and a normal control that were digested with the restriction endonuclease EcoR1 and probed with a radiolabeled RARalpha probe. The left lane is from an M3 patient while the right lane is a normal control. The same is true for the gel on the right except the probe is complementary to the PML gene. In both examples the normal control expresses only one version of either the PML or RARalpha gene. This is, of course, the normal or wild-type allele. In contrast, the M3 patient expresses both the normal PML and RARalpha genes, as well as the fusion protein.40

Southern blot is by far the most cumbersome, labor intensive, and costly technique discussed but it provides the most flexibility of application. Through the development of `home grown' probes developed in research laboratories, unknown and even novel mutations can be detected by this technique for any given genetic disease.

Finally, reverse transcriptase PCR is not only able to detect the presence of the t( 15;17) translocation, but it is thought to be very useful in prognosing M3 patients. The reverse transcriptase PCR technique is used to detect mRNA in a cell to verify transcription of the gene in question. The mRNA is isolated, converted to DNA, and amplified using PCR, as illustrated in Figure 10a. Promyelocytes and blasts are obtained from the blood of M3 patients that present with leukocytosis or from the bone marrow of patients with leukopenia. The cells are lysed with detergent, and mRNA is isolated on an oligo dt (deoxythymidine) cellulose column. The poly adenosine tail, referred to as the poly A tail of eukaryotic mRNA, binds to the solid phase poly deoxythymidine resin in the column. Since tRNA and rRNA do not possess poly A tails, they flow through the column. The column is eluted and the mRNA is collected and reverse transcribed into complementary DNA known as cDNA. As illustrated in Figure 10b, primers are introduced that are complementary to the PML and RARa sequences that Flank the breakpoint of the cDNA molecules. The primers will bind the cDNA, and DNA polymerase will synthesize a new DNA strand. This synthesis occurs for multiple cycles between the primers, thus reproducing the fusion gene into millions of copies. The breakpoint sequence that was amplified by the PCR technique is then electrophoresed and reacts with ethidium bromide in the gel. Ethidium bromide will intercalate into double stranded DNA, emit fluorescence, and can be visualized under LIV light. Fluorescence of the amplified sequence presumes the presence of the fusion mRNA from the patient and is considered a positive test.13

rtPCR is becoming useful in the detection of minimum residual disease in M3 patients who have already been treated. A striking correlation has been observed between the presence of fusion mRNA and relapse. Seventy-five percent of patients with a positive rtPCR experience relapse within six months of the positive test result. In contrast, only 6% of M3 patients with a negative rtPCR developed recurring disease within six months of testing.13 Since the mechanism of leukemogenesis appears to be directly correlated with the t( 15;17) translocation, the fusion mRNA is a good marker to determine prognosis and predict relapse. This technique is not as useful in other leukemias or cancers because nearly all cancers develop as a result of a multimutation process. Therefore, identification ofa single mutation, expressed with sufficient consistency on which to perform rtPCR to predict relapse, would be difficult in most forms of cancer.

MECHANISMS OF ATRA THERAPY

Because AML-M3 seems to be the result of a single and consistent genetic mutation, the development of a single therapeutic agent, such as ATRA, is much more feasible. Although the mode of action is not known, several mechanisms have been proposed. Recent evidence supports a mechanism involving ATRA induced release of the co-repressor complex from the fusion protein. The normal transcription repression complex (N-Cor-Sin3-HDAC) binds the PML/RARalpha fusion protein with sufficient affinity to prevent its release in the presence of physiological levels of retinoic acid. This is due to a reduction in sensitivity of the fusion protein to ATRA as compared to wild-type RARalpha. 17 Therapeutic levels of ATRA have been shown to bind the fusion protein and produce a conformational change in fusion protein structure.28 This would stimulate the dissociation of the fusion protein dimers from the repressor complex allowing the dimers to bind the histone acetylase transcription activator on the RA response elements (Figure 11), thus restoring transcriptional activation of RA controlled genes. 16,18-22,28 This mechanism is supported by the observation that HDAC1 (histone deacetylase) inhibitors, applied directly to M3 promyelocytes, will induce differentiation. Further evidence that this is the mechanism of leukemogenesis, involves the t(11;17) translocation occasionally observed in AML-M3. This is the only mutation described thus far that is resistant to ATRA induced differentiation therapy. It has been shown that the PLZF gene from chromosome 11 is also a transcription repressor, explaining the continued transcription inhibition of the RARalpha target genes under therapeutic concentrations of ATRA. However, direct application of HDAC1 inhibitors will also relieve the repression observed in the M3 promyelocytes expressing the t(11;17) translocation.41,42

As described earlier, protein dimerizations complicate the leukemogenic mechanism. In a similar manner protein/protein interactions may also be involved in the mechanism of ATRA induced differentiation therapy. It has been shown that the PML/RARalpha, fusion protein is transcribed to a much higher intracellular concentration than either PML or RARalpha. The implication is that the fusion protein would out compete and therefore displace the normal PML and RARalpha proteins from the RA response elements. This would result in less binding of the wild-type RARalpha, PML, and RXR to the RA response elements and a reduction in transcription. The combination of excess retinoic acid that converts the co-repressor to a co-activator, the freed retinoic acid response elements previously occupied by the fusion protein, and available RXR molecules previously sequestered by the fusion protein could all work together to reestablish normal mechanisms of myelocytic differentiation. 14,41,42

Several other mechanisms of ATRA function have been proposed. ATRA may also direct PML proteins to their proper location to reestablish PML function that may be important in differentiation. ATRA could also up regulate the non-rearranged RARalpha or increase the DNA binding affinity of RXR.43,44 Recent evidence suggests that ATRA therapy could enhance the natural degradation of PML/ RARalpha thus removing it from the protein competition pool. 29

FUTURE DIRECTION

Recent experimental evidence has shown a difference in binding affinities of the two major therapeutic isoforms of retinoic acid, ATRA and 9-cis-RA, to the three versions of the fusion protein dictated by the three PML breakpoints (bcrl, bcr2, bcr3) described earlier. The most common PML breakpoint occurs at bcrl and produces a fusion protein that expresses a greater sensitivity to ATRA than to 9cis-RA. In contrast, the bcr3 containing fusion protein is more sensitive to 9-cis-RA than the bcrl isoform or wild-type RARalpha. Since ATRA is the therapy of choice for M3 patients, this may, at least in part, account for the poorer prognosis of patients expressing the bcr3 fusion protein. In addition, this may suggest the possibility of specific differentiation therapy protocols (ATRA for bcr 1 and 9cis-RA for bcr3) in M3 patients based on the molecular identification of the fusion protein version acquired.44

The continued binding of the co-repressor complex by the fusion protein products of both the t(15;17) and t(11;17) translocations at physiological retinoic acid levels suggests a common leukemogenic mechanism. Although both oncoproteins contain RARalpha genes, the t(11;17) translocation is resistant to ATRA therapy. However, the histone deacetylase inhibitors are equally effective at restoring myeloid differentiation in both translocations. In addition, the recent observations that p300, CBP and MOZ are histone acetylases and involved in translocations observed in various cases of AML, further supports a common leukemogenic mechanism.45-49 This suggests a model of therapy using histone deacetylase inhibitors that may be effective for several types of AML by reversing the down regulation of the transcriptional repression machinery. This approach could effectively reverse blocks in differentiation across several types of AML regardless of the particular genes involved or the site of mutagenesis. Clinical trials using butyrate as a histone acetylase inhibitor are underway. Although insufficient clinical data are available to predict outcome, encouraging results have been observed in cell culture models.49

A novel approach to the treatment of M3 involving a laboratory engineered hammerhead ribozyme has been reported. The therapeutic agent is an RNAase designed to bind, digest, and inactivate PML/RARalpha mRNA. Preliminary results are promising, but as with any new therapy, much more work is needed to evaluate the efficacy of this form of treatment.36

TRAINING COURSE

A two-week course will be held on the campus of the Uniformed Services University of the Health Sciences, 4301 Jones Bridge Road, Bethesda, MD, from July 31 -August 11, 2000. The course will consist of a series of lectures and laboratory sessions

covering the diagnosis of parasitic infections of humans. Prior background in laboratory methods is recommended. The registration fee is $ 1000. Registration deadline: July 21, 2000.

Contact: Dr. John Cross or Ellen Goldman at the Department of Preventive Medicine and Biometrics at USUHS. (301) 295-3129/3139, (301) 295-1971 (fax).

REFERENCES

1. McKenzie SL. Textbook of hematology. 2nd ed. Baltimore: Williams and Wilkins; 1996. p 385-7.

2. Evans VJ. In: Stiene-Martin EA, Lotspeich-Steininger CA, Koepke JA, editors. Clinical hematology, principles, procedures and correlations. 2nd ed. Philadelphia: Lippincott-Raven; 1988. P 444-8.

3. Golomb HM, Rowley J, Vardiman J, and others. Partial deletion of the long arm of chromosome 17: a specific abnormality in acute promyelocytic leukemia. Arch Intern Med 1976;136:825.

4. Rowley JD, Golomb HM, Dougherty C. 15!17 translocation: a consistent chromosomal change in acute promyelocytic leukemia. Lancet 1977;1:549. 5. Lavau C, Dejean A. The t(15;17) translocation in acute promyelocytic leukemia. Leukemia 1994;8:(suppl 2) S-9.

G. Borrow J, Goddard AD, Sheer D, and others. Molecular analysis of acute promyelocytic leukemia breakpoint cluster region on chromosome 17. Science 1990;249:1577.

7. de The' H, Chomienne C, Lanotte M, and others. The t(15;17) translocation of a acute promyelocytic leukemia fuses the retinoic acid receptor a gene to a novel transcription locus. Nature 1990;347:558.

8. Alcalay M, Zangrilli D, Pandolfi PP, and others. Translocation breakpoint of acute promyelocytic leukemia lies within the retinoic acid receptor oc locus. Proc Natl Acad Sci USA 1991;88:1977.

9. Lemmons R, Eilender D, Waldmann R, and others. Cloning and characterization of the t( 15;17) translocation breakpoint region in acute promyelocytic leukemia. Genes Chromosomes Cancer 1990;2:79.

10. de The H, Lavau C, Marchio A, and others. The PML-RARa fusion mRNA generated by the t(15;17) translocation in acute promyelocytic leukemia encodes a functionally altered RAR Cell 1991;66:675.

11. Kakizuka A, Miller WH, Umesono K, and others. Chromosomal translocation t(15;17) in human acute promyelocytic leukemia fuses RARa with a novel putative transcription Factor, PML. Cell 1991;66:663.

12. Grignani F, Ferrucci PF, Testa U, and others. The acute promyelocytic leukemia-specific PML-RARa fusion protein inhibits differentiation and promotes survival of myeloid precursor cells. Cell 1993;74:423.

13. Diverio D, Pandolfi PP, Rossi V, and others. Monitoring of treatment outcomes in acute promyelocytic leukemia by RT PCR Leukemia 1994;8(7):1105. 14. Grignani F, Fagioli M, Alcalay M, and others. Review. Acute promyelocytic leukemia: From genetics to treatment. Blood 1994;83:10.

15. Lafage-Pochitaloff M, Alcalay M, Brunel V, and others. Acute promyelocytic leukemia cases with nonreciprocal PML/RARa or RAR(X/PML fusion genes. Blood 1995;85:1169.

16. Vyas RC, Frankel SR> Agbor P, and others. Probing the pathobiology of response to all-trans retinoic acid in acute promyelocytic leukemia: premature chromosome condensation/fluorescence in situ hybridization analysis. Blood 1996;87(1):218.

17. Guidez F, Ivins S, Zhu J, and others. Reduced retinoic acid-sensitivities of nuclear receptor corepressor binding to PML- and PLZF-RARa underlie molecular pathogenesis and treatment of acute promyelocytic leukemia. Blood 1998;91 (8):2634.

18. Leid M, Kastner P, Lyons R, and others. Purification, cloning and RXR identity of the HeLa cell factor with which RAR or TR heterodimerizes to bind target sequences efficiently. Cell 1992;68:377.

19. Collins SJ. Acute promyelocytic leukemia: relieving repression induces remission. Blood 1998;91 (8):2631.

20. Grignani F, Dematteis S, Nervi C, and others. Fusion proteins of the retinoic acid receptor-alpha recruit histone deacetylase in promyelocytic leukemia. Nature 391:815, 1998.

21. He LZ, Guidez F, Tribioli C, and others. Distinct interactions of PML-RARalpha and PLZF-RAR-alpha with co-repressors determine differential responses to RA in APL. Nat Genet 18:126, 1998.

22. Lin RJ> Nagy L, Inoue S, and others. Role of the histone deacetylase complex in acute promyelocytic leukaemia. Nature 391:811, 1998.

23. Fagioli M, Alcalay M, Pandolfi P, and others. Alternative splicing of PML transcripts predicts coexpression of several carboxy-terminally difFerent protein isoforms. Oncogene 1992;7:1083.

24. Kalantry S, Delva L, Gaboli M, and others. Gene rearrangements in the molecular pathogenesis of acute promyelocytic leukemia. J Cell Physiol 1977;173:288.

25. Chang KS, Fan YH, Andreeff M, and others. The PML gene encodes a phosphoprotein associated with the nuclear matrix. Blood 1995;85:3646.

26. Koken MHM, Linares-Cruz G, Quignon F, and others. The PML growthsuppressor as an altered expression in human oncogenesis. Oncogene 1995;10:1315.

27. de The' H. Altered retinoic acid receptors. FASEB J 1996;10:955.

28. Schulman IG, Evans RM. Retinoid receptors in development and disease. Leukemia 1997;11(Suppl. 3):376.

29. Doucas V, Evans RM. The PML nuclear compartment and cancer. Biochim Biophys Acta 1996;1288:M25.

30. Dyck JA. Rapid diagnosis of acute promyelocytic leukemia by immunohistochemical localization of PML/RAR-alpha protein. Blood 1995;86:862.

31. Kastner P, Perez A, Lutz Y, and others. Structure, localization and transcriptional properties of two classes of retinoic acid receptor a fusion proteins in acute promyelocytic leukemia (APL): structural similarities with a new family of oncoproteins. EMBO J 1992;11:629.

32. Yang-Yen H, Zhang X-K, Graupner G, and others. Antagonism between retinoic acid receptors and AP-1: implication for tumor promotion and in

flammation. New Biol 1991;3:1206.

33. Weis K, Rambaud S, Lavau C, and others. Retinoic acid regulates aberrant nuclear localization of PML/RARoc in acute promyelocytic leukemia cells. Cell 1994;76:345.

34. Perez A, Kastner P, Sethi S, and others. PML/RAR homodimers: distinct DNA binding properties and heteromeric interactions with RXR. EMBO J 1993;12:3171.

35. Dyck J, Maul G, Miller VC; and others. A novel macromolecular structure is a target of the promyelocyte-retinoic acid receptor oncoprotein. Cell 1994;76:333.

36. Early E, Dmitrovsky E. Acute promyelocytic leukemia: retinoic acid response and resistance. J Invest Med 1995;43(4):337.

37. Pemrick SM, Lucas DA, Grippo JF. The retinoid receptors. Leukemia 1994;8(Suppl.3):Sl.

38. Waters JJ, Barlow AL, Gould CP Demystified - FISH. Mol Pathol 1998;51 (2):62.

39. Werner M, Wilkins L, Aubele M, and others. Interphase cytogenetics in pathology: principles, methods, and applications of fluorescence in situ hybridization (FISH). Histochem Cell Biol 1997;108:381.

40. Ream VI; Field KG. Molecular biology techniques: an intensive laboratory course. San Diego: Academic Press; 1999. p 163-89.

41. Fenaux P, Chomienne C, Degos L. Acute promyelocytic leukemia: biology and treatment. Sem in Oncol 1997;24(1):92.

42. Chomienne C, Ballerini P, Balitrand N, and others. Normal RARa gene transcripts are modulated during RA induced granulocytic differentiation in APL cells. J Clin Invest 1991;88:2150.

43. Degos L, Dombret H, Chomienne C, and others. All-trans retinoic acid as a differentiating agent in the treatment of acute promyelocytic leukemia. Blood 1995;85(10):2643.

44. Benedetti L, and others. Characterization of the retinoid binding properties of the major fusion products present in acute promyelocytic leukemia cells. Blood 1997;90(3):1175.

45. Sobulo OM, Borrow, Tomek R, and others. MLL is fused to CBP, a histone acetyltransferase, in therapy-related acute myeloid leukemia with a t(11;16)(q23;p13.3). Proc Natl Acad Sci USA 1997;94:8732.

46. Taki T, Sako M, Tsuchida, and others. The t(11;16)(q23;p13) translocation in myelodysplastic syndrome fuses the MLL gene to the CBP gene. Blood 1998;98:3945.

47. Borrow J, Stanton V, Andresen JM> and others. The translocation t(8;16) (pI 11;p13) of acute myeloid leukaemia fuses a putative acetyltransferase to the CREB-binding protein. Nature Genet 1996;14:33.

48. Carapeti M, Aguiar RC, Goldman JM, and others. A novel fusion between MOZ and the nuclear receptor coactivator TIF2 in acute myeloid leukemia. Blood 1998;91;3127.

49. Wang J, Saunthararajah Y, Redner RL, and others. Inhibitors of histone deacetylase relieve ETO-mediated repression and induce differentiation of AMLI-ETO leukemia cells. Cancer 1999;59:2766.

Tim R Randolph MS is an assistant professor in the Department of Clinical Laboratory Science, School of Allied Health Professions, Saint Louis University Health Sciences Center, St Louis MO.

Address for correspondence: Tim R Randolph, Department of Clinical Laboratory Science, School of Allied Health Professions, Saint Louis University Health Sciences Center, 3437 Caroline St, St Louis MO 631041111. (314) 577 8518, (314) 577 8503 (fax). Randoltr@ slu. edu

Copyright American Society for Clinical Laboratory Science Spring 2000
Provided by ProQuest Information and Learning Company. All rights Reserved