The current therapy for glioblastoma multiforme involves total surgical resection followed by combination of radiation therapy and temozolomide. Unfortunately, the efficacy for such current therapy is limited, and newer approaches are sorely needed to treat this deadly disease. We have recently described the isolation of bacterial proteins and peptides with anticancer activity. In phase I human clinical trials, one such peptide, p28, derived from a bacterial protein azurin, showed partial and complete regression of tumors in several patients among 15 advanced-stage cancer patients with refractory metastatic tumors where the tumors were no longer responsive to current conventional drugs. An azurin-like protein called Laz derived from
The brain is a protected organ and therefore limits the number of compounds that can enter it from peripheral circulating blood. Thus only selected compounds such as glucose, alcohol, nicotine, and others can enter the brain to help nourish it or affect it in other ways, but many other blood components cannot enter the brain. This physical barrier, termed the blood-brain barrier (BBB), is characterized by monolayers of endothelial cells that are tightly packed to prevent leakage of brain components or entry of nonpermissive substances [
The presence of the BBB, and the efficient efflux mechanism, has significantly limited the number of drugs that can enter the brain to treat brain pathologies, including brain tumors. Transmembrane diffusion, usually by small lipophilic molecules, has been the major route to designing drugs that can enter the brain in amounts sufficient to provide effective treatment, although such molecules can be targets of P-glycoprotein efflux [
The entry limitations to the brain of most drugs intended for the treatment of brain pathologies have triggered a flurry of activities to design drugs, lipophilic compounds with polar groups, for example, that can cross the BBB to enter the brain parenchyma by disrupting the tight junctions of the endothelial cells of the brain capillaries. One approach uses convection-enhanced delivery of the drugs through insertion of selected catheters containing drugs that penetrate the interstitial space to enter the brain parenchyma. Another development involves the use of recombinant adeno-associated virus expressing neurotrophic factors [
The transcytosis of a carrier protein through binding with the ApoB LDL receptor binding domain of about 38 amino acids has prompted investigations to use a family of peptides derived from proteins that utilize LRPs for crossing the BBB to deliver therapeutic agents [
Glioblastoma multiforme (GBM) is one of the most deadly gliomas because of its high genetic diversity, a complex vasculature giving rise to intratumoral pressure and invasiveness, and the lack of access of most drugs because of the presence of entry barriers including the BBB. The invasive nature of GBM, allowing it to spread throughout the CNS, makes the therapy all the more difficult, and the prognosis remains poor [
As mentioned above, only selected molecules such as glucose, insulin, and ethanol can cross the BBB to reach the brain parenchyma and many blood components cannot cross this barrier, yet certain extracellular bacterial pathogens that are basically normal inhabitants in the pharyngeal region (
An important aspect of basic research is the unpredictability of the direction it will take. Thus fundamental basic research has contributed to many out-of-the-box ideas and new approaches for drug development and therapeutic intervention. Let us give an example. About 12 years ago, we had no interest or expertise in cancer research, including development of anticancer drugs. We were studying the mechanism of infection of a bacterium known as
Highly encouraged by these initial observations that we have found a new toxin against macrophages (and a new hitherto unknown function for azurin), our group checked the cytotoxic activity of purified azurin against primary mouse macrophages isolated from the peritoneum. To our utter surprise, azurin again demonstrated high cytotoxicity in J774 cells but very little cytotoxicity for mouse primary peritoneal macrophages. Repeated experiments showed the same results. With a great deal of confusion and anguish, we searched for the difference between J774 cells and normal peritoneal macrophages, until it dawned to us that J774 macrophage cell line is derived from tumors, allowing them to grow rapidly in cell culture and facilitating their use as macrophages.
Does azurin demonstrate cytotoxicity against tumor cells but not against normal cells like peritoneal macrophages? Further experiments clearly demonstrated that azurin not only had high cytotoxicity against human cancer cell lines such as melanoma and breast cancer showing very little toxicity in normal cells, but also showed a great deal of promiscuity in attacking and demonstrating high cytotoxicity against viruses such as the AIDS virus HIV-1 or parasites such as the malarial parasite
Azurin, 128 amino acid long and about 14 kDa, is not only produced by
Occurrence of azurin-like proteins across a phylogenetic tree. By homology searches, azurin orthologous were found in a variety of bacterial species members of the gamma- and betaproteobacteria but absent from the other bacterial phyla and the Eukarya. Black and white boxes indicate presence and absence of azurin, respectively. The box with vertical bars indicates the existence of an azurin-like protein (termed Laz) in
Of interest is the presence of an azurin-like protein, termed Laz, uniquely found in
(a) A multiple amino acid alignment of 9 representative bacterial azurins and 3 azurin-like proteins (Laz) from neisserial species. Identical residues (
Azurin is believed to be a weapon produced by some pathogenic bacteria such as
Does the presence of the H.8 epitope in azurin, that is, Laz, allow facilitated disruption of any entry barrier to the glioblastomas or even help in crossing the BBB? Even more pertinent is the question: does the bacterial weapon azurin work in the human environment to attack tumors and cause their regression?
Because azurin and Laz are bacterial proteins, for preclinical or human clinical trials, they face stringent regulation and may have toxic cellular contaminants. On the other hand, a peptide derived from azurin or Laz, which can be chemically synthesized as a drug, has entry specificity in cancer cells and demonstrated anticancer activity and therefore can be tested clinically for toxicity and efficacy in humans. Such a peptide is p28, a 28-amino-acid peptide derived from azurin (azurin 50–77) with entry specificity in cancer cells, as well as with anticancer activity [
List of issued US patents on protein drugs azurin and Laz.
Title | Inventors | Patent number | Date of issuance |
---|---|---|---|
Cytotoxic factors for modulating cell death | Chakrabarty AM, Das Gupta TK, Punj V, Zaborina O | 7,084,105 | August 1, 2006 |
Compositions and methods for treating HIV infection with cupredoxins and cytochrome c | Chakrabarty AM, Das Gupta TK, Yamada T, Chaudhari A, Fialho A, Hong CS | 7,301,010 | November 27, 2007 |
Compositions and methods for treating malaria with cupredoxin and cytochrome | Chakrabarty AM, Das Gupta TK, Yamada T, Chaudhari A, Fialho A, Hong CS | 7,338,766 | March 04, 2008 |
Compositions and methods for treating conditions related to ephrin signaling with cupredoxins | Chakrabarty AM, Das Gupta T, Yamada T, Chaudhari A, Fialho A, Zhu Y | 7,381,701 | June 03, 2008 |
Cytotoxic factors for modulating cell death | Chakrabarty AM, Das Gupta TK, Punj V, Zaborina O, Hiraoka Y, Yamada T | 7,491,394 | February 17, 2009 |
Compositions and methods for treating HIV infection with cupredoxin and cytochrome c | Chakrabarty AM, Das Gupta, T, Yamada, T, Chaudhari A, Fialho A, Hong CS | 7,511,117 | March 31, 2009 |
Compositions and methods to control angiogenesis with cupredoxins | Mehta RR, Taylor BN, Yamada T, Beattie CW, Das Gupta TK, Chakrabarty AM | 7,556,810 | July 07, 2009 |
Compositions and methods to prevent cancer with cupredoxins | Das Gupta TK, Chakrabarty AM | 7,618,939 | November 17, 2009 |
Cupredoxin derived transport agents and methods of use thereof | Chakrabarty AM, Das Gupta T, Yamada T, Fialho A | 7,691,383 | April 06, 2010 |
Compositions and methods for treating malaria with cupredoxin and cytochrome | Chakrabarty AM, Das Gupta T, Yamada T, Chaudhari A, Fialho A, Hong CS | 7,740,857 | June 22, 2010 |
Transport agents for crossing the blood-brain barrier and into brain cancer cells, and methods of use thereof | Hong CS, Yamada T, Fialho A, Das Gupta TK, Chakrabarty AM | 7,807,183 | October 5, 2010 |
The fact that out of 15 patients where the tumors were refractory to all conventional drugs, 2 patients showed complete regression of their tumors with enhanced life expectancy raises important questions not only about the unique mode of action of p28, and therefore azurin, but also about the genetic traits, either in the tumor genome or in the patient genome, that make these tumors highly susceptible to p28. The complete tumor regression may also be due to the right concentrations of p28 and the right length of time of treatment. It should be noted that p28 is only a part of azurin, and there are other domains in azurin, p26, for example, that demonstrate anticancer activity through inhibition of receptor tyrosine kinases such as EphB2 [
The presence of an additional H.8 epitope in neisserial azurin (Laz) begs the question if indeed the H.8 epitope facilitates the anticancer weapon azurin to cross any entry barrier to glioblastomas, and given the surface exposure of Laz in
Left row: structural depiction of azurin, Laz (with the H.8 epitope in the N-terminal), and azurin with the cloned H.8 epitope in its N-terminal. The area corresponding to the p28 peptide region is marked. Middle row: the entry of fluorescently labeled azurin, Laz, and H.8-azurin in breast cancer MCF-7 and glioblastoma LN-229 cells is shown. The red color reflects the Alexa-Fluor-568-conjugated protein while the blue color represents the nucleus stained with DAPI. Right row: cytotoxicity of azurin, H.8-azurin, and Laz at 3 different concentrations towards LN-229 glioblastoma cells. The detailed methodologies have been described by [
How may the H.8 peptide facilitate such entry of azurin in glioblastoma cells? Does it disrupt any entry barrier, perhaps the tight junctions in the LN-229 cells? To address such questions, we used chemically synthesized 39-amino-acid H.8 peptide by itself, labeled with Alexa Fluor 568, for its entry or unlabeled H.8 peptide for cytotoxicity in LN-229 and MCF-7 cells. The H.8 peptide by itself did not show any entry or cytotoxicity in these cancer cells [
Does the H.8 epitope promote the crossing of the BBB as well? Azurin, H.8-azurin, and Laz were each fluorescently labeled with green fluorescing IR dye 800 CW (LI-COR Biotechnology, Lincoln, Nebraska) and 500
Odyssey scanning of brains from mice previously injected peritoneally with green fluorescent IR-dye-conjugated Laz, H.8 azurin, and azurin (from
Surgery, radiation, and chemotherapy, the main approaches to cancer therapy, do not work very well for glioblastoma mainly for two reasons. Firstly, few chemotherapeutic drugs can cross the BBB to reach the brain tumor in significant amounts for therapeutic purposes. A second problem involves the high invasive nature of glioblastomas within the central nervous system, greatly accelerating the chances for relapse of the tumor. Thus any approach to glioblastoma therapy must take into consideration not only the need for complete tumor removal but also to reduce the possibility of a relapse. It is interesting to point out that azurin has both these properties. Azurin and p28 not only have anticancer activity, interfering in multiple steps in cancer growth, but also have cancer preventive properties, as determined by inhibition of the oncogenic transformation of normal mouse mammary cells to develop precancerous lesions in presence of a potent carcinogen, 7,12-dimethyl-benz-anthracene [
Finally, the efficacy of p28 in allowing complete regression of tumors in a melanoma and a sarcoma patient, where the tumors were nonresponsive to conventional drugs, raises some interesting questions. Apart from the question of the genotypic characteristics of the tumor or patient genomes as mentioned earlier, could it be due to the unique characteristics of the bacterial protein/peptide drugs that work through novel multiple pathways? It should be noted that p28, and in fact azurin, are known to enter breast cancer tumors very efficiently and allow significant tumor shrinkage
Cytotoxicity effect of AT-01 peptide and cisplatin at 1
The authors acknowledge financial support from the Department of Biotechnology, Government of India to Amrita Therapeutics. This study was supported in part by Fundação para a Ciência e a Tecnologia (FCT), Portugal (Grant PTDC/EBB-BIO/100326/2008).