Campisi seems to regard the tiny human fibroblast cells she studies as fellow-creatures that she has gotten to know well-individuals with a birth, a childhood, a growth pattern, dangers and misadventures along the way to maturity, and-in the fullness of time-old age and a death that may or may not be preprogrammed to happen when it does.
Campisi doesn't believe in doing biology with statistical analysis of huge, anonymous populations of cells. `'If you need statistics to do molecular biology, you need a better experiment," she says. Instead, she and her coworkers in LBL's Life Sciences Division study small populations of cells-no more than 100 or so at a time-and sometimes fol low the fortunes of an individual cell from birth to death. To get this intimate, they rely on tools like a precision-con trolled needle that can inject material directly into the nucleus of a living cell without hurting it.
In the last few years, Campisi has been exploring what happens when cells reach old age-a biological phenom enon known as cellular senescence. She is particularly interested in the relationship between cellular senescence and the familiar stages of growth, aging, disease, and death as they are experienced by multicellular living organisms (our selves, for instance).
With her team of students, technicians, and postdoctoral fellows, Campisi is trying to understand the genetic and molecular basis of senescence at the cellular level-in particular, to identify the genes associated with senescence and the mechanisms through which they work.
A gene is the unit of DNA that carries the instructions for making a single protein; proteins are the substances that carry out virtually all the biological processes of the body. It is estimated that each human cell contains about 100,000 different genes. Because these are present in every cell in the body, any cell, theoretically, could produce any protein. However, cells are specialized to perform certain functions, and most cells are able to `'turn on" (express) no more than about 10 percent of the genes they contain. Some of those genes-no one knows how many or which ones-must be involved in the universal processes of aging.
Senescence, says Campisi, is a fundamental feature of normal cells. It happens to most differentiated cells, even primitive ones, both in the living animal and when grown in culture.
Usually, cells that can divide in the body are also capable of proliferating in culture. However, after a number of populations (usually about 50), there is a progressive decline in proliferative capacity, and, finally, a complete end to cell division. (Only a few primitive stem cells-such as the inner cell mass of the early embryo- may have an unlim ited proliferative potential in culture. )
In addition to the cessation of growth, cells which have entered senescence generally display a number of changes in form and function. These may include an increase in size, alterations in subcellular architecture and extracellular matrix, and changes in the synthesis of some proteins. Senescence, though a distinct stage in cellular life, is not death, or anything like it. Senescent cells can still perform many biological functions and sometimes take on new ones. The only thing they lose- permanently, it appears-is the ability to divide and, thus, to make new cells. In culture, senes cent cells may live for months or even years before random processes slowly diminish their number.
While cellular senescence is nearly universal, the degree to which it occurs depends on the species. Humans are particularly affected. `'Immortal" cells (those having an unlimited lifespan in culture) very rarely arise spontaneously from human cell cultures. This is not the case for some other species, such as rodents. When rodent cells are grown in culture, senescence is sometimes circumvented because immortal cells, while unusual, do occur, and can multiply and eventually dominate the culture.
Despite the universality and inevitability of cellular senescence in human beings, its significance is not well un derstood. What is its biological role? Does it have a value to the organism-for example, as a defense against the prop agation of cancer-causing mutations that often produce immortal cells? And what relationship does the fate of individual cells have to that of the organism of which they are a part? That relationship is not a simple one: for example, skin biopsies from even very old human individuals still contain plenty of proliferative, nonsenescent cells. It is doubt ful that cells reach the end of their proliferative lifespan very often in you and me.
On the other hand, the changes in form and function that cells undergo with senescence could result in at least some of the features associated with aging-for example, the wrinkling of the skin. And, while human cells in cultures normally become senescent after about 50 divisions, cells that were taken from aged adults reach senescence in half that time. These observations have led to the view that senescence in cell cultures reflects processes that occur during aging in the living human being.
Back at Boston University Medical School, where she headed a research team before coming to LBL two years ago, Campisi was initially interested in cancer research-particularly in studying the molecular and genetic events that transform a normal cell into a cancerous one. She now believes that `'proto-oncogenes"-genes that play a vital role in human growth but can also trigger cancer when they undergo mutations or are expressed abnormally-are deeply in volved m normal aging as well.
`'One could say that senescence is the opposite of cancer," says Campisi. `'Some of the same genes that drive un controlled proliferation in cancer seem to be under tight control or actually turned off in senescence. We know that in cancer, a proto- oncogene becomes out of control. Perhaps in senescence, a proto-oncogene becomes repressed."
Campisi and a colleague at Boston University, Monica Peacocke, made an important discovery. They succeeded in showing that a particular proto-oncogene, known as fos, is repressed in senescent cells and cannot be reactivated by any stimulus-though in nonsenescent younger cells the same gene can readily be turned on and off. At LBL, Campisi and her team- including Goberdhan Dimri, David Rosen, and Akif Uzman-are following up this discovery with studies aimed at identifying the genes and gene products responsible for senescence and the closely related state called "terminal differentiation."
In terminal differentiation, a cell has achieved the ideal form for carrying out its biological function. Once a cell is terminally differentiated, it will not start to divide and proliferate again no matter how it is stimulated. Campisi is inclined to believe that there is a close connection between senescence and terminal differentiation. If this view is cor rect, the loss of the ability to divide may be just one of many far-reaching changes in form and function that occur as a cell matures into the final phase of its life.
In the human body, cells fall into three classes. Some-like nerve and muscle cells-become terminally differen tiated while the organism is still an embryo. Such a cell, if damaged later on in life, will never rejuvenate itself; instead, a new, previously undeveloped nerve cell may mature and take the place of the damaged one.
Other cells-like those in the lining of the stomach and colon-divide continuously until the organism's death puts a stop to all life processes. Such vigor may seem desirable, but it also carries a risk, since every division presents an opportunity for a potentially cancer-causing mutation to enter the genetic material of the cell and its progeny.
Finally, a large group of human cells seem to move easily back and forth between periods when they divide (pro liferation) and periods when they do not divide (quiescence). Given time, however, such cells may gradually lose the ability to switch from quiescence back to proliferation and are said to have entered senescence.
A major group of such cells are the fibroblasts. Fibroblasts are cells found in many organs of the body. Often, they form a kind of support structure for more specialized cells. Thus, in the skin, fibroblasts underlie the horny epidermal cells like a cushion. Most of the time, these fibroblasts are fully functional-for example, they produce collagen-but they are quiescent; they do not divide or grow, although they retain the capability to do so if appropriately stimulated.
One way this can happen is through wounding. When skin cells are damaged or destroyed, new ones must be cre ated and brought to maturity in a hurry. The underlying fibroblasts, like an army roused from bivouac, seem to get the message right away. Growth factors, released at the site of wounding, carry the message to the nuclei of surrounding cells, which respond by switching on a large number of specialized genes.
Among the genes that become activated are `'proliferation genes"-genes whose principal function seems to be to take cells out of the quiescent stage and initiate cell division. (Other types of genes are also switched on by wounding- for example, ones that produce the enzymes and other proteins needed for cell repair.)
The proliferation genes are not numerous (about 100 different types have been identified so far) and are remark ably uniform for the various organ systems in the body; that is, the genes that control the proliferation of liver cells often perform the same function for lung cells. Some, though not all, of the proliferation genes appear to be proto- oncogenes.
It has been estimated that every cell in the body has a one in a million chance of undergoing a mutation in one of its genes. However, very few mutations are cancer- causing. In order for a mutation to lead to tumor formation, it must occur either in a proto-oncogene, in one of the genes that control the expression of proto-oncogenes, or in another class of genes called tumor- suppressor genes, which may counteract the action of some proto-oncogenes. Furthermore, the mutation must not be a lethal one, or it would simply kill the cell without spreading the effect.
Because we have trillions of cells in our bodies, if a single oncogenic mutation were enough to trigger tumor growth, cancer would be much more common than it is. Fortunately, the alteration of more than one proto-oncogene or tumor-suppressor gene seems to be required for cancer to occur.
In the healthy individual, proto-oncogenes are kept in check by two fundamentally different mechanisms. One in volves external control mechanisms like growth factors or the extracellular matrix. The other relies on tumor-suppres sor genes.
Less than a dozen of these tumor- suppressor genes have been identified. Says Campisi, `'We don't know much about these genes, but we think that they seem to keep the cell from runaway growth. And it seems reasonable that some tumor-suppressor genes are also responsible for the establishment and maintenance of cellular senescence."
Campisi's research using human fibroblasts provides firm evidence for this view. While still at Boston University, Campisi and her coworkers were able to demonstrate that the proto-oncogene fos is turned off when cells become se nescent. Now, in more recent work at LBL, her team has discovered that a tumor-suppressor gene known as Rb prob ably acts to keep fos turned off.
The function of the Rb gene, first discovered in connection with certain eye tumors in children, has long been a subject of speculation. Because it can be expressed by every cell in the body, it seems likely that it has global functions. According to Campisi's new findings, one of those functions may be that it suppresses the proliferation gene fos after the onset of senescence. Since fos is involved in regulating the expression of differentiated products in fibroblasts, such as some enzymes that degrade collagen, its repression could also contribute to an outward sign of aging: the wrinkling of skin. Says Campisi, `'Our findings suggest that fos repression may be one of the hallmarks of a global change in cell function."
Can fos be turned back on once it has been repressed by Rb? Another recent result from Campisi's lab-involving the oncogene of a nonhuman tumor virus known as T antigen, or TA-provides a clue. Campisi's team has discovered that TA reverses the repression of fos in senescent fibroblasts. TA has long been recognized as the only gene product that can produce what appears to be a brief and incomplete attempt at growth in a senescent cell. In earlier studies, TA had been found to induce senescent human fibroblasts to begin DNA synthesis, although the cells do not continue to grow, and the effect is short-lived. Campisi's work clearly shows that one of the ways TA does this is by turning fos expression back on.
In other recent work, Campisi's team has identified another gene, ID, that seems to act like or, that is, it is repressed in senescent cells, it cannot be turned on again with growth factors or other stimuli, and its repression is also probably mediated by the tumor-suppressor gene Rb. However-and here the story gets even more intriguing-ID is not a pro to-oncogene but a gene that controls terminal differentiation.
Another promising finding involves a gene associated with a protein `'kinase"- an enzyme that changes the func tion of proteins produced by other genes by adding a phosphate group. This gene, known as cdc2, also turns out to be repressed in senescent cells. Campisi's postdoctoral fellow, Goberdhan Dimri, has succeeded in cloning this gene, in cluding the portion of the gene that controls its expression. The availability of these clones should help Campisi and her team to unravel further the mysteries of senescence.
Campisi has noticed that a lot of people-fellow-scientists and laymen alike- take a personal interest in studies of aging, even in the senescence of the lowly fibroblast. `'We're all getting older," she says, `'and we want to know what's going on." Particularly, people want to know if her research sheds any light on the great unanswered question, `'Has nature programmed us to die?"
''There is a lot of debate about this," says Campisi. Some scientists point to the fact that although the average life span has risen in recent times through disease prevention, medical intervention, and improved health habits, the upper limit (about 120 years for humans) has probably not changed since biblical times.
`'This would seem to argue for a built-in senescence," says Campisi. However, some evidence points the other way. `'There have been studies in fruit flies, for example, in which maximum life span was extended significantly, even dou bled, using the modern techniques of molecular genetics to slow the rate of aging," she says. `'So we have to say that the case is still undecided, and the questions of whether and how we're programmed to die remain open for now."
-JUDITH GOLDHABER