This interview was updated 3/13/06 to correct typographical errors.
Head of hematopathology
Harbor-UCLA Medical Center and Los Angeles Biomedical Research Institute
At A Glance
Name: Beerelli Seshi
Position: Head of hematopathology, Harbor-UCLA Medical Center and Los Angeles Biomedical Research Institute; associate professor of pathology, David Geffen School of Medicine at UCLA, since 2002.
Background: Associate professor of pathology and oncology, attending hematopathologist, H. Lee Moffitt Cancer Center & Research Institute, University of South Florida, 1996-2002.
Assistant professor of pathology and laboratory medicine, assistant director of hematology laboratory unit, University of Rochester School of Medicine and Dentistry, 1990-1996.
Staff pathologist and medical director of hematopathology and flow cytometry, Madigan Army Medical Center, 1989-1990.
Residency in anatomic pathology and clinical pathology, Yale-New Haven Hospital, 1984-1988.
MBBS Osmania Medical College, Hyderabad, India, 1977.
The National Institutes of Health recently awarded Harbor-UCLA Medical Center researcher Beerelli Seshi a $609,318 grant to study leukemic marrow stromal cells using functional proteomics. ProteoMonitor spoke with Seshi to find out more about his background and his current research project.
What is your research background? How did you get into using proteomic techniques to study leukemia?
I have an MD background. I did my residency at Yale … in pathology. I'm a board-certified pathologist. And for 15 years, I have been an academic clinical hematopathologist specializing in leukemia/lymphoma diagnosis. That's what I do for my bread and butter.
When the leukemia patient presents, bone marrow is usually completely loaded with leukemia cells — completely replaced. You look at the marrow, then you make the diagnosis, and the hematologist treats the patient, and dramatically all the leukemia would be gone, and a few more weeks later normal marrow is back. It's a pretty fascinating experience to look at.
When you look at the marrow, it has a very interesting architecture. There are lots and lots of different types of cells. It's almost like a Los Angeles community — lots of heterogeneity. All the blood cells are produced there throughout our life — red blood cells, white cells, and platelets. Each one of them has their own counterparts in the bone marrow. So morphologically, when you look at it, it's a very complex environment. But what you don't see when you look at the marrow histology under the microscope are what are called stromal cells, because they are mesenchymal stem cells. They are few and far between — it's a kind of infrequent cell type.
However, there are well-established culture systems. In two weeks, with a marrow sample, you can grow tons of these cells. These fibroblast-like cells have the capacity to sustain the growth of hematopoietic stem cells. So when you seed these cultures with hematopoietic stem cells, they survive because they adhere to the stromal fibroblastic cells and they survive because these stromal cells provide the necessary adhesion molecules and growth factors.
People have been studying this for the past 25 years. But somehow, people thought the stromal system, which looks kind of heterogenous, with different types of morophologies, was a mixture of muscle cells, bone cells, fat cells, fibroblasts, etc. That is what people thought made up a nutritive cocktail that is needed for preservation of hematopoietic stem cells.
Then when my lab wanted to pinpoint what these cell types are, we discovered that there are not five different cell types. There is only one cell type with all the makeup in one cell.
That means stromal cells simultaneously express markers for muscle cells, bone cells, fat cells, fibroblasts. So there is only one cell type, including neuronal markers. So it dawned upon us there is only one cell type, contrary to what has been thought. This we called mesenchymal progenitor cells.
My lab has studied these cells, using a variety of cytochemical, immunocytochemical, electron microscopic methods. Subsequently, we microdissected a single isolated stromal cell, and did single-cell microarray analysis using Affymetrix [arrays]. We showed that transcripts belonging to muscle cells, bone cells, fat cells, fibroblasts — they're all there within the same cell. So that way we have characterized the system to our satisfaction.
And later on, we found that these cells, when they are co-injected into SCID mice, they support the engraftment of hematopoietic stem cells, and also prevent graft vs. host disease.
In transplantation, there are two issues: that the transplant is not grafting, and that the cells are acting against the host — that's called graft vs. host disease. So these cells take care of both issues.
I'm coming to the leukemia part. But before you can study the proteomics, you need to know what the heck you are looking at. Because if you think there are five different cell types, and you do the proteomic studies, you won't know where these proteins are coming from. It would be extremely complex to be able to analyze.
Now that we have identified the cell, and we can purify these cells, the issue is that these cells not only support the growth and preservation of the normal hematopoietic cells, but also the leukemia cells. So probably the same adhesion molecules, the same growth factors produced by these stromal cells, influence the growth and proliferation of these leukemic cells. It's from the same source, because bone marrow is the place where leukemia arises.
So it's very critical to constantly look at both sides — the normal and leukemic sides.
Is that what your current research project is focused upon?
Now the question is how to go about understanding the molecular mechanisms of stromal cells. My lab has been interested in this for 15, 16 years.
Initially, we tried to understand by doing a kind of cell blotting. What we were doing was taking the bone marrow stromal proteins, doing either 1D or 2D electrophoresis, then blotting the proteins onto a PVDF or nitrocellulose membrane. Then we blocked the background with BSA or whatever, just as in Western blotting. And then we added cells — hematopoietic or leukemic cells — to bind to the proteins that are separated by electrophoresis.
That system other people are applying. Basically, that's how my interest got started in proteomics. We wanted to move on to the global approach, which is the proteomic approach, where you can study thousands of proteins both in the normal and the diseased [samples].
We grow the stromal cells. We purify the cells from normal stroma and also from leukemic people. We isolate the proteins and do the proteomic analysis.
The grant I got was in two phases. For the first phase we demonstrated that different techniques are applicable. We used both iTRAQ and DIGE with three fluorescent dyes.
These initial studies fulfilled the required milestones. That's why I got this recent grant awarded to do more systematic application of these technologies to understand proteomic changes between normal and leukemia-associated stromal cells.
The stromal cells make up the microenvironment. Other labs have shown that stromal cells support the adhesion of the leukemic cells, and also prevent the apoptosis of the leukemic cells, which means they're facilitating their survival and growth. People also have shown that if leukemic cells bind to stromal cells, they become drug resistant. So there is a tremendous interest in dissecting out the stromal cell proteomic make up.
It's possible that these studies might lead to diagnostic markers and therapeutic targets. These are buzz words, but that's the ultimate goal.
Why not study the leukemia cells directly?
Well, there are a lot of people who are studying them. But microenvironment has been neglected. By the neighborhood you live in, one would be able to tell something about you. If you're in a posh neighborhood, that means you make this much money, you do this, you do that, whereas if you live in a ghetto, it's different. So by studying the microenvironment you would be able to gain insight that would not be possible just by studying the leukemic cells themselves. As somebody said, 'Know thy neighbor.'
And these studies may have implications for solid tumors as well. For example, there are breast cancer cells. But breast has stroma, too. Ovary has stroma, too. They are all epithelial tumors — that means carcinoma — but their growth is promoted by the surrounding stromal cells. We don't know whether those stromal cells are identical or similar to the marrow stromal cells, but nevertheless, they make up the microenvironment.
What kind of proteins have you found through your cell-blotting experiments?
We identified a few bands, and they were showing nice adhesion. Then we ran out of funding, and that work did not proceed further. But recently somebody has applied this technology and identified a molecule that contributes to the hematopoietic niche of the bone marrow. So that's the one that attracts the hematopoietic cells to the bone marrow.
These things go in incremental fashion. I can tell you how hard it was for the past two years to meet the milestones. You tend to overpromise when you write a grant without realizing all the difficulties. But nonetheless we did it.
Did you find different proteins using high-throughput proteomics compared to what you found using cell blotting?
We still haven't done any comparative analysis, but using iTRAQ technology, we identified more than 900 proteins. And we haven't fractionated the samples yet. So this is kind of a first glimpse of the proteome of the stromal cell, because to my knowledge, this is the first systematic investigation of the proteome of the stromal cell.
Our goal was not to dwell on one particular subproteome, like membrane, or cytosolic — that would be our next step. But we simply catalogued 900 proteins. That in itself is a feat as far as we're concerned.
On the other side, using the DIGE technology, we have been able to detect more than 5,000 protein spots with spectacular resolution.
What have you found in comparing normal vs. diseased stromal cells?
We found about 70-some proteins by iTRAQ technology, and about 150 or so by DIGE technology. But these are from limited numbers of samples, so this is simply a way of establishing the technology. They are of importance, but I don't know if they're of statistical significance.
We haven't discovered a therapeutic target or diagnostic marker yet. It's a long-term process. It takes time. But I think I have a very important system characterized. I have my work cut out for the next 20 years.
Does this project include validation?
Yes. For the next two or three years, that's what we'll be doing. We'll focus on one or two leukemias. These first few years were just to establish the technology.
For validation, first and foremost we need to study more samples, and to do statistical analysis. I may have to zero in on certain subproteomes, like membrane proteins, etc. Then, the interesting proteins we may have to confirm by Western blots and by immunocytochemistry. And that's not enough. Then you may have to do some other studies, like RNAi, or knockout mice. It's a long-term process. We are just at the very beginning.
What are the current methods for diagnosing leukemia?
First you look at the morphology. Based on the morphological appearances, we have a clinical flow cytometry lab and we do immunohistochemistry and flow cytometry on a routine basis for making the diagnosis. That's the mainstay of diagnosis at this point.
So you look under the microscope, look at an aspirate smear, and see if there's anything abnormal. Then you send some sample for cytogenetic analysis, and some for flow cytometry. We have 40 to 50 markers for different cell lineages, and based on those you determine what type of leukemia it is. That's all we have at this point.
How long is this NIH grant for?
It's technically for two years. Next year I would presumably be getting the same amount. I may also take the stem cell angle and hope to get some funding from California State Stem Cell Program. There are multiple options.
Basically, I'm most excited by the biology. Technology is only a means to an end. This bone marrow to me is as fascinating as the brain, or some other organ. And I wouldn't say this about other tissues.
I believe I have a worthwhile problem on which to spend my life. Hopefully, we'll make important discoveries.