Harlow K. Fischman, Ph.D.

College of Physicians and Surgeons, Departments of Medical Genetics, New York State Psychiatric Institute, 722 West 168th Street, New York, N.Y. 10032, and Genetics and Development, College of Physicians and Surgeons, Columbia University

Dennis D. Kelly, Departments of Behavioral Physiology, New York State Psychiatric Institute, and Psychiatry, College of Physicians and Surgeons, Columbia University

Correspondence to: Dr. Harlow K. Fischman, P.O. Box 6025, Englewood, CO – 80155-6095, Voice Mail (800) 228 – 8193, Mail Box #23422.

We thank Dr. Donald Ross for his statistical analyses and John D. Rainer of the New York State Psychiatric Institute for his helpful ideas and support.  We also thank Dr. Mohammed Osman for his cooperation and support; and Emilia Moralishvili, Dr. Osafradu Opam, and Dr. Ludmilla Skaredoff for their technical assistance. Correspondence should be addressed to H. K. Fischman. Dennis D. Kelly is deceased.

We have previously established that acute psychogenic stress in rats induces genetic damage on both the chromosomal and molecular levels (Fischman, Pero, and Kelly, 1996).  Rats subjected to stress showed increases in both the level of Sister Chromatid exchanges (SCEs) and chromosome aberrations (CAs) in bone marrow cells.  The increases, to differing degrees, in SCEs and CAs induced by the exposure of rats to a variety of stressors, such as cold and warm water swim, white noise, and continuous or intermittent foot-shock, demonstrated that this is a general phenomenon of stress.  These stressors differed from each other both quantitatively  and qualitatively.  The varying experimental paradigms followed, demonstrated that such damage can occur in as short a time as 2 hrs, and endure for at least 25 hrs following exposure to stress.  Furthermore, the detection of stress-induced damage by means of Unscheduled DNA Synthesis extended these observations to the molecular level and to yet another cell type, leucocytes (Fischman, et al., 1996).

A focus on the role of stress in disease has led to the development of the field of Psychoneuroimmunology.  Intensive research in this field in recent years has substantiated that there are physiological and molecular, as well as anatomical connections between the Central Nervous System (CNS), and the endocrine and immune systems (Kropiunigg, 1993; Maier, Watkins, & Fleshner, 1994).  For example, the classical experiments of Riley (1975), demonstrated that exposure to or protection from stress can respectively speed up or slow down development of mammary tumors in mice carrying the Bittner oncogenic virus.  Other research has demonstrated that psychological factors, such as stress, contribute to the predisposition, onset, and course of various illnesses, such as depression, infections, rheumatoid arthritis, coronary heart disease, and cancer in humans, and to herpes simplex, poliomyelitis, Coxsackie B, polyoma, and induction and growth of Walker carcinoma and Ehrlich ascites in animals (Dorian and Garfinkel, 1987; Eysenck, Grossarth-Maticek, and Everitt, 1991).  Our hypothesis is that a parallel or related situation exists with regard to stress and the Genetic system.  In the three experiments to be described in this paper, we continued, on the chromosomal level, a systematic exploration of some aspects of psychogenic stress which may effect this system.  We have developed an In Vivo bone marrow technique for examination of SCEs (Fischman, et al., 1996).  SCEs are microscopically detectable interchanges between the replicated chromatids of metaphase chromosomes that are associated with repair of damaged DNA (Latt, 1979).  Exposure to mutagen/carcinogens results in a dose-dependent elevation of SCEs, and when repair-deficient cells are treated with these agents, large increases in SCEs result (Latt, 1979; Perry and Evans, 1975).  Although the mechanism by which SCEs are produced has not yet been elucidated, detection of elevated SCEs has proved to be an exceptionally sensitive measure of the mutagenic potency of environmental agents in mammalian systems (Tucker and Preston, 1996; Wolff, Rodin, and Cleaver, 1977).

In the experiments to be described, we examined: the effect of  more prolonged periods of stress; the possible role of the endocrine system; and the relationship between stress and chemical mutagens.  (1)  Our original studies focused on very short periods of stress, ranging from 3 – 20 min, those usually considered “acute”.  We determined to expand our findings by examining the effects of more extended periods of stress.  If stress periods are lengthened, is damage continued, enhanced, or decreased? This experiment examines whether repeated exposures to the same stressor results in a gradual decline in the genotoxic properties of  stress, as do other bodily responses to stress, such as adrenal activation, which show adaptation( Bodner, Kelly, Brutus, & Glusman, 1978).  A comparison of the levels of SCEs and of CAs in groups of rats exposed to different durations of stress, also examines whether the genotoxic properties of stress display a different time course of adaptation to chronically stressful conditions than do many other bodily functions affected by stress.  We chose to scrutinize extended stress periods of 3 and 10 days.

(2)  The endocrine system plays an integral role in mediating the effects of the nervous and immune systems (Vollhardt, 1991). In order to explore whether it takes an analogous or identical part with respect to the nervous and genetic systems, we examined whether hormones secreted by glands in the  hypopothalamic-pituitary-adrenal axis have an impact on stress-induced chromosome damage.  This was accomplished by an examination of the effects of foot-shock stress on SCE and CA levels in hypophysectomized(Hypox) rats.

(3)  Physical agents, such as UV and ionizing radiation, and many chemical agents, act as mutagens (Bloom, 1981).  Many mutagens increase SCE and CA levels(Tucker and Preston, 1996).  In light of our demonstration of the SCE- and CA-inducing actions of psychogenic stress, the ubiquity of physical and chemical mutagens in our environment (Bloom, 1981), and of stress in our society (Kiecolt-Glaser & Glaser, 1987), it is appropriate to raise the question as to the possibility of interaction between stress and mutagens.  We examined this prospect by exposing rats to both a chemical mutagen, Mitomycin C (MMC), and foot-shock stress.


In all experiments, the subjects were male albino Sprague-Dawley rats.  They were individually housed with continuous access to food and water, and maintained on a 14 hr light/10 hr dark cycle, with the exception of the Prolonged Continuous Stress experiment.  Whenever the experimental design permitted, all stress sessions, as well as the time of sacrifice, occurred + or – 1 hr from the midpoint of the light cycle. Two hours following a single exposure, or immediately following the last of a series of stress sessions, each animal was lightly anaesthetized with ether, and a Bromodeoxyuridine (BrdU) pellet (250 mg/Kg B.W.) was implanted subcutaneously on the back of the neck.  Twenty-one hours after BrdU implantation, the rat was injected intraperitoneally with Colcemid (0.6 mg/Kg B.W., i.p.).  Two hours later the rats were sacrificed by guillotine, and hindbone bone marrow preparations were made for analysis of SCEs (Allen, Shuler, & Latt, 1978).  Slides were stained according to the method of Goto, Akematsu, Shimazu, & Sugiyama (1975), for permanent fluorescence plus Giemsa (FPG) SCE preparations.  Fifty well-spread and well-stained, apparently unbroken second division (after BrdU addition) metaphases from each rat were selected for SCE analysis.  Only cells with less than 39 chromosomes were excluded (2n = 42 in R. Rattus).  In experiments in which chromosome aberrations were also to be analyzed, 100 first division metaphases from the same slides were selected.  Slides were coded and scored blind.  SCEs were scored as the mean number of SCEs/cell and chromosome aberrations were scored as the mean number of cells with at least one chromosome break in 100 cells (Fischman, et al., 1996).

Prolonged, Continuous Stress  (PCS) 

Ten rats, 11-12 weeks of age, bred and maintained as previously described, and equated for weight, were assigned to three groups, which were subjected to respectively: No-Stress (NS), n = 4; 72-hr Stress (72S), n = 3; and 240-hr Stress (240S) n = 3.  Throughout the stress condition the subject was exposed to constant dim illumination in the test chamber.  Food and water remained freely available.  The stress paradigm consisted of  alternating 30 min sessions of white noise stress and Conditioned Emotional Response (CER) training.  The initial session was always white noise.  White noise was delivered by two independent audio noise generators over two loud speakers mounted 5-10 cm from the Plexiglas test cage in which the subject was housed.  In turn, this cage was enclosed within a larger ventilated, sound-insulated chamber so as minimize the influence of external stimuli and maximize the efficiency of the speaker system generating the white noise.  Combined white noise output was calibrated at 120 db.  Free field noise intensity measurements were made at the completion of each experiment.  CER sessions consisted of audio warning stimuli which were intermittently paired with brief inescapable footshocks according to the following schedule.  The onset of each minute in the CER session had a 50% probability of triggering a warning stimulus.  The duration of the pure tone warning stimulus was fixed at 40 secs.  The offset of each warning stimulus carried a 50% probability of causing a 500-msec shock (2 mA constant current) to be delivered through the grid floor of the test chamber.  The polarity of the grid bars were “scrambled” at a high speed to prevent intensity-attenuating escape responses within the 500-msec duration.  The long-term probabilities of the CER protocol were as follows:  On the average there were 15 warning stimuli per 30 in CER session.  Hence, at 40-s per stimulus, the average subject spent approximately 10 mins of each CER session in the presence of warning stimuli and 20 mins in safety.  With the probability of shock set at P = 0.5 per warning stimulus, the long-term expectancy of shock was 7.5 per 30-min CER session or 180 per 24 hrs.  Although the actual shock frequency was low, the threat of shock was constant and high, for the subject could not predict which minute would be associated with a warning stimulus, and which warning stimulus would be paired with a terminal shock, except in a probabilistic manner.  However, because no free shocks were programmed to occur unless proceeded by a warning tone, the latter was a predictor of shock and hence acquired the properties of a conditioned emotional stimulus.


Seventeen 58 day old rats were hypophysectomized or sham-operated 9 days prior to the start of the experiment. Their age at sacrifice was 67 days.  Animals were divided into 4 groups: Hypophysectomy-Stress (HS),  n = 5; Hypophysectomy-No Stress (HC), n = 4; Sham-operated-Stress (SS), n = 4; and Sham-operated-No Stress (SC), n = 4.  Stressed animals were subjected to 240 Intermittent Foot-Shocks (IFS).  These were administered  as 2.5 mA, 60 Hz constant current pulses of 1-s duration repeated every 5 s for 20 min.  The non-stressed groups (HC and SC) were handled for 3.5 min.  Slides were prepared and scored for SCEs and CAs in the same manner as in the prolonged continuous stress experiment.


Twenty rats, bred and maintained as previously described, and equated for weight, were assigned to five equal groups, which were subjected to respectively: No Stress (NS), No Stress plus Saline (NSS), No Stress plus Mitomycin-C (NSMC), Stress plus MMC (SMC), and Stress plus Saline (SS). All animals except the NS group were injected subcutaneously with either 10-8 M (f.c.) of MMC, an amount determined in preliminary experiments to approximately double the SCE level, or an equal volume of saline.  The SMC and SS groups were then subjected to IFS as previously described for the hypophysectomy experiment.  This intensity and duration of foot-shock has been shown to increase the SCE level in male rats (Fischman, et al., 1996).  Two hours after injection, or handling in the case of the NS rats, BrdU pellets were subcutaneously implanted.


Prolonged ,Continuous Stress

Both SCEs and chromosome aberrations were subjected to One-Way Analysis of Covariance (ANCOVA).  For chromosome aberrations, P values were derived from one-tailed t Tests :  Overall ANOVA:  F = 15.2, df = 2,7, P<0.005.  For SCEs, P values were derived from two-tailed t Tests :  Overall ANOVA: F = 20.7, df = 2,7, P<0.005.  72 hr and 240 hr Stress groups showed significant and highly significant increases respectively in both SCEs and chromosome aberrations (see Table 1).

SCEs:   The level for the 72 hr Stress group was significantly higher than for the No-Stress group (P<0.05).  SCEs for the 240 hr Stress group were elevated in a highly significant manner over those of the No-Stress group (P<0.0005).  The level for the 240 hr Stress group was significantly higher than that of the 72 hr Stress group (P<0.01).

Chromosome aberrations:  The level for the 72 hr Stress group was significantly higher than for the No-Stress group (P<0.05).  Aberrations for the 240 hr Stress group were elevated in a highly significant manner over those of the No-Stress group (P<0.001).  The level for the 240 hr Stress group was significantly lower than that of the 72 hr Stress group (P<0.05).

SCEs/CAs:  The correlation was extremely weak ( R = 0.185).


Both SCEs and chromosome aberrations were subjected to One-way Analysis of Covariance (ANCOVA).  P values were derived from two-tailed t Tests.  Overall ANOVA: F = 14.3. df = 3, 28, P<0.0005.

SCEs:   IFS elevated the level in both Stressed Sham-Operated  (P<0.0001) and Stressed Hypoxed rats (P<0.0005), compared with their No-Stress controls.  However, there was no significant difference between the No-Stress, Sham-operated and No Stress, Hypoxed groups (P<0.67) or between the Stressed, Sham-Operated and Stressed, Hypoxed (P<0.550 groups.  Overall, the elevation of SCEs in the combined Stressed groups was highly significant when compared with the combined No-Stress groups (P<0.0001).  There was no significant difference between the combined Sham-Operated groups  and the combined Hypoxed groups (P<0.91).  There was a significant difference between groups that differed in both respects, Hypox, No-Stress compared with Sham-Operated, Stress, (P<0.0005). (see Table 2)

Chromosome aberrations:   These results paralleled those found with respect to SCEs.  IFS elevated the level in both Stressed Sham-Operated (P<0.0001) and Stressed Hypoxed rats (P<0.0001) compared with their No-Stress controls.  There was however, no significant difference between the combined No-Stress, Sham-Operated and No-Stress, Hypoxed groups (P<0.62), or between the combined Stressed, Sham-Operated and Stressed, Hypoxed groups (P<0.054).  Overall, the elevation of chromosome aberrations in the combined Stressed groups was highly significant when compared with the combined No-Stress groups (P<0.0001).  Chromosome aberrations in the Hypoxed, Stressed group were elevated in a highly significant manner over the Sham-Operated, No-Stress group (P<0.0001).  There was no significant difference between the combined Sham-Operated groups and the combined Hypoxed groups (P<0.054).  There was a significant difference between groups that differed in both respects, Hypox, No-Stress compared with Sham-Operated, Stressed (P<0.0005). (see Table 2).


SCEs were subjected to One-way Analysis of Variance (ANOVA).  P values were derived from one-tailed t Tests.  Overall ANOVA:  F = 8.4, df = 4, 15, P<0.001.  There was no significant difference in SCE level between the Home Cage group (NS) and the No-Stress, Saline-injected group  (P<0.15).  There were significant differences between the Home Cage and both the Stress + Saline and No-Stress, MMC groups, which had higher SCE levels (P<0.01 and P<0.0001 respectively).  There was a highly significant difference between the Home Cage and Stress + MMC group, which had a higher SCE level (P<0.0005).  The Saline-injected group compared with the remainder of the groups, all of which had higher SCE levels, in the following manner:  vs Stress + MMC, a highly significant difference (P<0.005),  vs Stress + Saline and vs No Stress, MMC, no significant difference (both groups, P<0.5).  Stress + Saline was compared with No-Stress, MMC, which although it had a higher SCE level, showed no significant difference (P = 0.72).  The Stress + Saline group was significantly different from Stress + MMC, which had a higher SCE level (P<0.05), and Stress + MMC’s SCE level was significantly higher than that of No Stress, MMC (P<0.05) (see Table 3).


In many circumstances the physiological and behavioral effects of acute stress have been shown to have different and even paradoxical effects from those produced by prolonged or chronic stress. (Galinowski, 1993; MacLean, Walton, Wenneberg, Levitsky, Manderino, Waziri, Hillis, & Schneider, 1997 ).  In our original studies, subjects were exposed only to a brief, single occurrence of a stressful situation.  However, unlike initial exposure to stress, repeated exposure to the same stressor is known to result in a progressive decline in certain behavioral responses to stress, such as post-stress analgesia (Bodner, et al., 1980).  Stated alternatively, most stress-activated systems in the body eventually develop tolerance to those environmental demands or situations with which they have become familiar through experience.  Consequently, it would seem important to our line of research and to our cytogenetic system to compare the effects of acute and prolonged exposure to a known SCE-inducing stressor in order to determine whether stress-induced cytogenetic damage also shows adaptation.  A number of investigators have examined the effects of stress on tumor induction (Sklar and Anisman, 1981).  Although it is difficult to draw general conclusions from studies in this field because of the use of different systems, stressors, and measures of tumor growth, it nevertheless appears that acute stress enhances tumor growth while chronic stress inhibits it.  This differential response was observed using a wide variety of stressors, such as restraint, electroconvulsive shock, swimming, and noise.  To add to the complexity of this situation, it is even a matter of debate what constitutes chronic stress.  We extended our original observations to 3 and 10 day periods in order to cover periods of stress usually considered chronic in rats.

The data on SCEs comports well with our original findings.  SCE levels in unstressed animals in the PCS experiments, as well as in the Hypox and Mutagen/Stress

experiments, were consistent with those found over a number of years in our laboratory, and which have demonstrated a remarkable consistency.  In a total of 41 animals, the mean level has been 2.62 +/- 0.06 SEM SCEs/Metaphase.  The increase in SCE level in both the 72 and 240 hr PCS experiments demonstrates that the elevation observed during acute stress (Fischman, et al., 1996) also occurs during these longer stress periods.  In addition, the  SCE level was higher in 240 hr stress than in 72 hr stress, and thus may indicate either a dose response-type reaction, mediated by number of anticipated stresses, or an accumulation of SCEs over time.  SCE persistence depends on many factors: tissue type, nature of the inducing agent, number of times the animal was treated with the agent, duration of  exposure, what part of the cycle the cell was in during exposure, and how many times the cell divided prior to analysis (Tucker, Strout,

Christensen, & Carrano, 1986).  In absolute terms, the 240 hr SCE level of 5.01 SCEs/cell was in the same range as the highest obtained in our original experiments (6.36 SCEs/cell with long intermittent foot-shock).  In relative terms, the 240 hr SCE level was approximately double that of the controls compared with 2 1/2 X in the aforementioned foot-shock experiment. (Fischman, et al., 1996)  The situation with respect to CAs however, is different.  Although their level also increased in both 72- and 240-hr stress,  it was higher in the 72-hr experiment.  Thus, it would appear that adaptation occurred with respect to CAs but not with SCEs.  There is one report that indicates that adaptation to psychogenic stress may protect against the effects of a chemical mutagen.  Meerson and his co-workers adapted mice to moderate periodic hypoxia and repeated electric pain stresses of moderate intensity.  Treatment of unadapted animals with Dioxidine induced CAs in 11% of bone marrow cells, whereas adapted mice apparently had less than half this amount (Meerson, Kulakova, & Saltykova, 1993).  One possible explanation for the disparity between CAs and SCEs in the PCS experiment would be that the mechanisms for the induction of SCEs and CAs, in so far as they are presently understood, are different, especially those aspects which impact their accumulation and duration.  For example, researchers have reported the persistence of SCE levels, days and even weeks subsequent to a single exposure to a chemical mutagen(Tucker, et al., 1986).  The persistence of CAs in rat bone marrow cells may be of shorter duration.  If adaptation occurred sometime during the 240-hr stress, there may have been a loss or repair of chromosomes with breaks although during the same period SCEs persisted.

There may be other factors besides the phenomenon of adaptation, that make PCS different from acute stress with respect to chromosome damage.  Laudenslager and his colleagues showed that Concanavalin A-induced lymphocyte proliferation in rats is suppressed by inescapable, but not escapable shock (Laudenslager, Ryan, Drugen, Hyson, & Maier, 1983). This suggests that the ability of an organism to exert at least partial control over a stressor may be an important parameter of the degree to which stress interacts with cellular processes.  This line of research might be fruitfully pursued in future investigations of the effects of stress on the genetic system by analyzing genotoxic endpoints, using the “learned helplessness” paradigm.

Two obvious differences from our original experiments on acute stress are that the number of stress events have increased and that the time during which the rats are actually being stressed and during which they are under threat of stress has also increased.  Experiments, in which these factors are varied, may further elucidate their roles. The shock intensity in the PCS experiment (2 mA) differed from that used in the Hypox and Mutagen/Stress experiments (2.5 mA).  Thus, the PCS experiment is not comparable to the Hypox and Mutagen/Stress experiments with respect to shock intensity, and we do not know if these two intensities differ in genotoxicity.  A 14 hour light and 10 hour dark cycle was routinely used in the Authors’ laboratory.  The sole exception was that of the PCS experiment, during which the shock chamber was continuously dimly lighted for the duration of the procedure.  It is not possible to definitively state that there was no stress associated with this schedule because the controls in the PCS experiment were not exposed to constant dim illumination.  However, the control animals were essentially comparable to the controls in all other stress experiments performed in our laboratory over a number of years.  This is demonstrated by the consistency of their SCE levels (2.75 SCEs/metaphase).   

            In the search for mechanisms by which psychogenic stress is converted into chromosome and DNA damage, the endocrine system is a prime candidate, mainly due to its role in mediating between the nervous and immune systems. The Hypothalamus (HT) is known to be involved in stress reactions.  Corticotrophin Releasing Factor (CRF), produced in the HT, stimulates the Anterior  Pituitary to produce ACTH, which, in turn, induces the Adrenal Cortex to produce glucocorticoids (GCC).   GCCs may be involved in genotoxic processes, although convincing evidence for this has yet to be elucidated.  In our Hypox experiment, the lack of difference, with respect to genotoxicity, between sham-operated rats and those which had their pituitaries removed, would seem to preclude a role for glucocorticoids in this process. Nevertheless, because there is a possibility that GCCs can be induced without the intervention of ACTH, we have examined the literature for evidence of genotoxicological effects of GCCs.

Veien and Wulf (1980), indicated that GCCs may play a role in chromosome loss in leucocyte cultures from Sarcoidosis patients. It seems at least equally likely that Sarcoidosis itself, a chronic progressive systemic granulomatous reticulosis of unknown etiology, may have brought about the loss.  Sinues et al., in 1992, reported on a prospective study on asthmatic patients receiving glucocorticoids in combination with Theophylline (TP) and beta-adrenergics.  An increase in SCEs was observed.   Most likely, this was caused by the TP for the following reasons: (1) There was no significant difference between the GCC-treated patients and other subgroups, and; (2)  Methylxanthines, the chemical class to which TP belongs, are known inducers of chromosome alterations.  Tedeschi and his colleagues (Tedeschi et al., 1993) showed that recombinant human growth hormone (rhGH) raises the level of SCEs in Bleomycin-treated leucocyte cultures from otherwise normal children receiving therapy for short stature.  rhGH also induced chromosome rearrangements in these cells.  In an interesting paper, Joseph-Lerner et al. (1993), demonstrated that SCE frequency varied with menstrual cycle time, and had a positive correlation with human chorionic gonadotrophin, testosterone and FSH, and estradiol levels.  Finally, Dhillon and Dhillon (1996) showed that Norethisterone, a commonly used, long-acting, injectable oral contraceptive, induces SCEs and micronuclei in vivo in mice and in cultured human leucocytes.

The Autonomic Nervous system has been shown to modulate the effects of stress on the Immune System.  This is most likely accomplished through direct innervation of the major lymphoid organs and through catecholamine receptors on leucocytes, which produce functional changes in immunological cells (Dhillon and Dhillon (1996); Friedman and Irwin, 1997).  Stress-induced genotoxic damage may be brought about through similar or identical pathways.  Hypothalamic CRF, acts on the Sympathetic Nervous System as well as on the HPA axis.  This results in increased levels of corticosteroids, catecholamines, and certain opiates (Friedman and Irwin, 1997).

The results of our Hypox experiment were clarifying, but not definitive.  Both measures of chromosome damage, SCEs and CAs, were elevated by foot-shock stress, as had been observed in our original experiments. Although there was some difference in the level of CAs in hypoxed and sham-operated stressed animals, it did not reach the level of statistical significance.  We conclude therefore, that there was no difference between the Hypox and Sham-operated groups. However, it must be kept in mind that hypophysectomy is an area fraught with many conflicting results. For example, some aspects of stress-induced analgesia are attenuated by hypox while others are enhanced (Kelly, 1982).  The results in the present study do not eliminate the possibility that other parts of the endocrine and neuroendocrine systems may be involved.  Experiments, involving deficits in other hormones, would be appropriate avenues of  further inquiry.  If correlations between such deficits and stress-induced chromosome damage are found, it could be determined whether hormone replacement therapy restores the normal  situation.

In the mutagen/stress experiment, we showed that both foot-shock and MMC individually elevated SCEs. When animals were exposed during approximately the same time period to both the chemical mutagen and to psychogenic stress, the effect on SCE level was the sum of the individually-induced levels.  Thus, it appears that, at least with respect to this particular mutagen, neither interference nor synergism is involved, but instead there is an additive effect.  However, the effects of saline injection complicate the situation.  Although there was no statistically significant difference between the Home Cage controls and the Saline-injected groups, the SCE level induced by saline injection was sufficiently elevated so that there also was no significant difference between the No Stress, Saline and the No Stress, MMC groups or between the Stress, Saline and Stress, MMC groups.  We suspect that it is not the saline itself, but the handling and injection of the animals that acts as yet another genotoxic stress.  Indeed, it was similar observations, in a prior investigation by the senior author, which eventually stimulated this line of inquiry into the genotoxic properties of stress.  During an examination of the long-term effects of heroin on Rhesus monkeys, it was noticed that SCE levels in saline-injected controls, as well as in the heroin-exposed monkeys, had risen significantly (Fischman, Roizin, Moralishvili, Albu, Ross, & Rainer, 1983).  Unlike the heroin-habituated animals, these controls underwent a great deal of stress for 6 months, due to the necessity of restraining them for daily injections.

A logical step in further investigation of genotoxic stress would be to inquire in what ways stress acts like mutagens.  One obvious direction would be to see how much genetic damage stress can cause.  We have been able to induce SCE levels as high as 6.36/cell (Fischman, et al., 1996), an elevation equivalent to those induced in mice by 9.0 X 10-7M Mitomycin C (Allen and Latt, 1976).  The large body of epidemiological data collected on so-called stress-related disorders have indicated that stress plays a role in the pathogenesis of  many diseases.  The importance of such a role is, up to now, conjecture (Dorian & Garfinkel, 1987).  Finding the highest levels of chromosome damage that can be induced by stress in experimental animals would give an estimate of the mutagenic potential of stress, compared to that of known mutagens, and thus, would give a better idea of the relative effects of chemical mutagens and genotoxic stress.

In such an investigation, the question would arise, as to whether more than one aspect of stress plays a role in the damage.  Each putative stressor could be experimentally dissected to examine such aspects as intensity, duration, pattern, adaptation, inescapability, and unpredictability.  Further investigation should attempt to discern whether there are stress equivalents of the dose response effect of chemical mutagen concentration.  There may be a “dose”-related increase in SCE level as some parameters of stress increase.  If such exist, they might be useful in titrating the effects of various types of stressful situations.  Such a phenomenon could be useful in an experimental paradigm, and cellular mechanisms involved in stress-related disorders might be analyzed conveniently in a laboratory animal model.  Two methodological assumptions underlying this approach are: (1) that exposure of the organism to behavioral stress is analogous to the presence of chemical mutagens in terms of the nature of the genotoxic damage induced, although not necessarily in terms of a common mechanism, and (2) that it is appropriate to utilize SCE levels as a “biological dosimeter” of this damage.

There have been a few reports of genotoxic damage in stressed animals.  Bone marrow cells of rats subjected to long-term stress after exposure to Cyclophosphamide, were reported to show an increase in mutagenic effect, with a higher frequency of translocation-like chromosomal interchanges, but not chromosome fragments, in rats from lines with a “high level of excitability” compared with those with “low-excitability” (Bykovskaia, Diuzhikova, Vaydo, Lopatina, & Shvartsman, 1994).  The functional state of the nervous system was measured by the response of the tibial nerve to electrical stimulation, after 15 days of daily stress caused by randomly presented pairing of light and unavoidable electric shock. This is claimed to give a general indication of the level of excitability of other parts of the nervous system (Vaydo, Shiryaeva, and Lopatina, 1993).  In a more recent report, an increase is described in CA rate in “highly excitable” rats when subjected to short-term emotional and analgesic stress. This was described as “low mutagenic” (Diuzhikova, Bykovskaia, Vaydo, Shiriaeva, Lopatina, & Shvartsman,  1996).   Thus, acute and chronic stress would appear to have paradoxical effects in inducing chromosome damage in rats with different thresholds of nervous system excitability.  One is tempted to make an analogy with different human personality types, which have been implicated in disease etiology (Fox, 1988), but in the absence of more solid evidence, such a conclusion is highly problematic.  Another possible interpretation is that the 15 days of stress used in distinguishing types of nervous system excitability, may have adapted the animals  to induction of CAs by that type of stress, but not to their induction by the acute stress, which apparently was not only quantitatively, but qualitatively different.

There are, as yet, few studies directly or tangentially linking stress-induced cytogenetic damage observed in laboratory settings, with similar findings in humans .  There is one intriguing report of increased SCE levels in soldiers deployed to Kuwait during the Iraq war The authors attribute this increase to hydrocarbons produced in oil fires.  However, they also speculate about other causes, such as psychological stress associated with being in harm’s way during a war.  (McDiarmid, Jacobson-Kram, Koloder, Deeter, Lachiver, Scott, Petrucelli, Gustavison, & Putman, 1995).  The SCE increase might also be caused by some combination of stress and mutagen action, such as exposure to N-Mustard gas (New York Times, 1996), similar to the mutagen-stress experiment described in the present study.  Another study reported a significant SCE increase in five volunteers after sleep deprivation (Bamezai and Kumar, 1992).  Sleep deprivation could be a psychological stress.  Morimoto and his colleagues found a higher level of CAs and SCEs in blood cells of people with “poor lifestyles”.  Not having too much perceived stress was included among the criteria for “healthy lifestyles”.  They state that cell cultures from males who have “good lifestyles” do not show as much increase in SCEs when treated with MMC as do cultures from men with “poor lifestyles”.  They report similar results for cultures treated with Ara-C, an inhibitor of radiation damage repair (Morimoto, Takeshita, Take-uchi, Maruyama, Ezoe, Mure, & Inoue, 1993).  In 1996, Silva and his co-workers, reported on SCE analyses of various categories of workers exposed to noise and vibration.  Among the groups studied, were helicopter pilots, and these exhibited high SCE frequencies.  The authors speculated that these results may not have reflected direct action of physical agents on DNA, but rather stress-induced pathophysiological alterations (Silva, Carothers, Branco, Dias, & Boavida, 1996).

Another line of investigation might examine a possible connection between the genotoxic effects of behavioral stress and those of heat-shock, until recently regarded as a phenomenon acting primarily on the physical and chemical level.  Heat-shock and behavioral stress may have a final common path.  It has been well-established that Heat Shock Proteins (HSPs) can be induced in bacteria and cultured mammalian cells subjected to toxic physical or chemical environments.  Psychogenic stresses in intact animals have also been found to induce HSPs in both brain and gut.  Restraint-water immersion stress of rats has been reported to have significantly increased the level of cerebral HSP70 mRNAs, perhaps indicating a protective role for families of HSPs in mammals under pychophysiological stress (Fukudo, Abe, Hongo, Utsumi, & Itoyama, 1995, Fukudo et al., 1997).  In addition, it has been reported that stressed rats which have been adapted, show a greater increase in HSPs than do stressed but unadapted animals (Meerson, Malyshev, and Zamotrinskii, 1993).  Meerson suggests that the antimutagenic effect of stress adaptation is likely to be accounted for by the stabilizing action of HSP (Meerson, Kulakova, & Saltykova, 1993).  However, Vamvacopoulos and his colleagues reported reduction of HSP90 in adapted rats (Vamvacopoulos, Fukuhara, Patchev, & Chrousos, 1993).  Spontaneously hypertensive rats and mice have abnormal HSP70, which is localized in the major histocompatability complex.  In contrast to adapted animals, these animals had low amounts of steady-state HSP70 (Hamet, 1992).  It is possible that the low concentration of HSP70 disposes them to be hypertensive.  Along this line of reasoning, it is interesting to note that CAs produced by 7, 12-dimethylbenz[a]anthracene (DMBA) in the bone marrow cells of hypertensive rats were reported to be three times above that of control rats (Ueda & Kondo, 1984).  A similar situation had previously been described by Pero and his associates, in a study of hypertensive men.  They demonstrated that N-acetoxy-2-acetylaminofluorene (NA-AAF)-induced UDS in lymphocytes showed a linear increase with blood pressure.  Most interestingly, NA-AAF-induced CAs also increased linearly with blood pressure (Pero, Bryngelson, Mitelman, Thulin, & Norden, 1976).

Thus, there is now a line of evidence indicating that a low level, or abnormal type, of HSP makes some animals more vulnerable to genotoxic damage, and linking stress, HSPs, and DNA damage with at least one condition, hypertension.


Allen, J. W. & Latt, S. A. (1976). In vivo BrdU-332258 Hoechst analysis of DNA replication kinetics and sister chromatid exchange formation in mouse somatic and meiotic cells. Chromosoma, 58(4), 325-340, November 29.

Allen, J. W., Shuler, C. F., & Latt, S. A. (1978). Bromodeoxyuridine tablet methodology for in vivo studies of DNA synthesis. Somatic Cell Genetics, 4(4), 393-405.

Bamezai, R. & Kumar, N. (1992). Sleep deprivation in human males and its effect on SCE rates in chromosomes – a preliminary study. Mutation Research, 283, 229-232.

Bloom, A. D. (1981). Guidelines for cytogenetic studies. In A. D. Bloom (Ed.). Guidelines for Studies of Human Populations Exposed to Mutagenic and Reproductive Hazards. pp. 3-35. White Plains, N.Y., March of Dimes Birth Defects Foundation.

Bodnar, A. G., Kelly, D. D., Brutus, M., & Glusman, M. (1978). Chronic 2-deoxy-D-glucose treatment: adaptation of its analgesic, but not hyperphagic properties. Pharmacology, Biochemistry & Behavior, 9, 763 – 768.

Bykovskaia, N. V., Diuzhikova, N. A., Vaydo, A. I., Lopatina, N. G., & Shvartsman, P. Ia. (1994). Frequency of chromosome aberrations induced by stress and cyclophosphane in bone marrow cells of rats selected for the threshold of nervous system excitability. Genetika, 30(9), 1224 -1228.

Dhillon, V. S. & Dhillon, I. K., (1996). Genotoxicity evaluation of norethisterone  acetate.  Mutation Research, 367(1), 1-10, January.

Diuzhikova, N. A., Bykovskaia, N. V., Vaydo, A. I., Shiriaeva, N. V., Lopatina, N. G., & Shvartsman, P. Ia. (1996). Frequency of chromosome aberrations induced by a single exposure to stress in rats selected for excitability of the nervous system. Genetika, 32(6), 851-853, June.

Dorian, B., & Garfinkel, P. E. (1987). Stress, immunity and illness – a review. Psychological Medicine, 17, 393-407.

Eysenck, H. J., Grossarth, M. R. & Everitt, B. (1991). Personality, stress, smoking, and genetic predisposition as synergistic risk factors for cancer and coronary heart disease. Integrative Physiological and Behavioral science, 26(4), 309-322, October- December.

Fischman, H. K.,  Pero, R., & Kelly, D. D. (1996). Psychogenic stress induces chromosomal and DNA damage. International Journal of Neuroscience, 84, 219 -227.

Fischman, H. K., Roizin, L., Moralishvili, E., Albu, P., Ross, D., & Rainer, J. D. (1983). Clastogenic effects of heroin in pregnant monkeys and their offspring. Mutation Research, 118(1 – 2), 77 -89, July.

Fischman, H. K. & Kelly, D. D., Unpublished.

Fox, B. H. (1988). Psychogenic factors in cancer. In S. Macs, C. D. Spielberger, P. B. Defares, & I. G. Sarason (Eds.). Topics in Health Psychology. pp. 37-55. New York: John Wiley & Sons.

Friedman, E. M. & Irwin, M. R., (1997), Modulation of immune cell function by the autonomic nervous system. Pharmacology and Therapeutics, 74(1), 27-38.

Fukudo, S., Abe, K., Hongo, M., Utsumi, A., & Itoyama, Y. (1995). Psychophysiological stress induces heat shock cognate protein (HSC)70 mRNA in the cerebral cortex and stomach of rats.  Brain Research, 675(1-2), 98-102, March 27.

Fukudo, S., Abe, K., Hongo, M., Utsumi, A., & Itoyama, Y. (1997). Brain-gut induction of heat shock protein (HSP)70 mRNA by psychophysiological stress in rats. Brain Research, 757(1), 146 -148, May 16.

Galinowski, A. (1993). Stress and panic. Immunologic aspects. Encephele, 19 Spec. No, 1, 147-151, March.

Goto, K., Akematsu, T., Shimazu, H., & Sugiyama, T. (1975). Simple differential Giemsa staining of sister chromatids after treatment with photosensitive dyes and exposure to light and the mechanism of staining. Chromosoma, 53,  223-230.

Hamet, P. (1992). Abnormal hsp70 gene expression: its potential key role in metabolic defects in hypertension. Clinical and Experimental Pharmacology and Physiology Supplement, 20, 53 -59.

Joeph-Lerner, N., Fejgin, M., Ben-Nun, I., Legum, C., & Amiel, A., (1993), The correlation between the frequency of sister-chromatid exchange and human reproductive hormones. Mutation Research, 300(3-4), 247-252, August.

Kelly, D. D. (1982). The role of endorphins in stress-induced analgesia. In K. Vereby (Ed.) Opioids in mental Illness: Theories, clinical observations and treatment possibilities (pp. 260-270), Vol. 398, Annals of the New York Academy of Sciences,  New York: New York Academy of Sciences.

Kiecolt-Glaser, J. K. & Glaser, R. (1987). Psychosocial moderators of immune function. Annals of Behavioral Medicine, (2), 16-20.

Kropiunigg, U. (1993). Basics in psychoneuroimmunology. Annals of Medicine, 25(5), 473-479, October.

Latt, S. A. (1979). Sister chromatid exchanges. Genetics, 92 (1 Pt. 1 Suppl.), 89-95, May.

Laudenslager, M. L., Ryan, S. M., Drugan, R. C., Hyson, R.L., and Maier, S. F. (1983). Coping and immunosuppression: inescapable but not escapable shock suppresses lymphocyte proliferation. Science, 221, 568-570.

MacLean, C. R., Walton, K. G., Wenneberg, S. R., Levitsky, D. K., Manderino, J. P., Waziri, R., Hillis, S. L., and Schneider, R. H. (1997). Effects of the Transcendental Meditation program on adaptive mechanisms: changes in hormone levels and responses to stress after 4 months of practice.  Psychoneuroendocrinology, 22(4), 277-295, May.

Maier, S. F., Watkins, L. R., & Fleshner, M. (1994). Psychoneuroimmunology: the interphase between behavior, brain, and immunity. American Psychologist, 49(12), 1004 -1017.

McDiarmid, M. A., Jacobson-Kram, D., Koloder, K., Deeter, D. P., Lachiver, R. M., Scott, B. G., Petrucelli, B. P., Gustavison, D., & Putman, D. (1995). Increased frequencies of sister chromatid exchange in soldiers deployed to Kuwait. Mutagenesis, 10(3), 263-265, May.

Meerson, F. Z., Kulakova, A. V., & Saltykova, V. A. (1993). The antimutagenic effect of adaptation to stress. Biulleten Eksperimentalnoi Biologii I Meditsiny, 116 (9), 292 – 295, September.

Meerson, F. Z., Malyshev, I. Iu., & Zamotrinskii, A. V. (1993). The generalized accumulation of stress proteins during body adaptation to stress exposures. Biulleten Eksperimentalnoi Biologii I Meditsiny, 116(9), 231-233, September.

Morimoto, K., Takeshita, T., Take-uchi, T., Maruyama, S., Ezoe, S., Mure, K., & Inoue, C. (1993). Chromosome alterations in peripheral lymphocytes as indices of lifestyle and genotoxicity. International Archives of Occupational and Environmental Health. 65 (I, Suppl.), 537-541.

New York Times (1996). Panel says chemical arms might be cause of gulf illness, November 21.

Pero, R. W., Bryngelsson, C., Mitelman, F., Thulin, T. & Norden, A. (1976). High blood pressure related to carcinogen-induced unscheduled DNA synthesis, DNA carcinogen binding, and chromosomal aberrations in human lymphocytes. Proceedings of the National Academy of Sciences, U.S.A., 73(7),  2496-2500, July.

Perry, P. & Evans, H. J. (1975). Cytological detection of mutagen- carcinogen exposure by sister chromatid exchange.  Nature, 258(5531), 121-125, November 13.

Riley, V. (1981). Psychoneuroendocrine influences on immunocompetence and neoplasia. Science, 212, 1100-1109.

Silva, M. J., Carothers, A., Branco, N. C., Dias, A., & Boavida, M. G. (1996). Sister chromatid exchange analysis in workers exposed to noise and vibration. Mutation Research, 369(1 – 2), 113-121, July 10.

Sinues, B., Broto, A., Suarez, M. A., Duce, F., Martinez-Berganza, A., & Bernal, M. L., (1992), Cytogenetic study in peripheral blood lymphocytes from asthmatic patients receiving continued therapy with theophylline. Mutation Research, 280(4), 271-277, October.

Sklar, L. S. & Anisman, H. (1981). Stress and cancer. Psychological Bulletin, 89(3), 369-406, May.

Tedeschi, B., Spadoni, G. L., Sanna, M. L., Vernole, P., Caporossi, D., Cianfarani, S., Nicoletti, B., & Boscherini, B., (1993), Increased  chromosome fragility in lymphocytes of short normal children treated with recombinant human growth hormone. Human Genetics, 91(5), 459-463, June.

Tucker, J. D., & Preston, R. J. (1996). Chromosome aberrations, micronuclei, aneuploidy, sister chromatid exchanges, and cancer risk assessment. Mutation Research, 365(1 – 3), 147-159, September.

Tucker, J. D., Strout, C. L., Christensen, M. L., & Carrano, A. V. (1986). Sister chromatid exchange induction and persistence in peripheral blood and spleen lymphocytes of mice treated with ethyl nitrosourea. Environmental Mutagenesis 8, 345-355.

Ueda, N. & Kondo, M. (1984). Chromosome aberrations induced by 7,12-dimethylbenz[a]-anthrocene in bone marrow cells of spontaneously hypertensive rats (SHR) and control Wistar Kyoto (WKY) rats: time course and site specificity. Journal of the National Cancer Institute, 73(2), 525-530, August 18.

Vamvakopoulos, N. C., Fukuhara, K., Patchev, V., & Chrousos, G. P. (1993). Effect of single and repeated immobilization stress on the heat shock protein 70/90 system of the rat: glucocorticoid-independent, reversible reduction of HSP90 in the liver and spleen. Neuroendocrinology, 57(6), 1057-1065, June.

Vaydo, A. I., Shiryaeva, N. V., & Lopatina, N. G. (1993). Divergent responses to long-term stressors in rat lines selected for different functional states of the nervous system. Behavior Genetics, 23(5), 499-503, September.

Veien, N. K. & Wulf, H. C. (1980), Chromosome analysis in patients with sarcoidosis. Archives of Dermatological Research, 268(1), 9-13.

Vollhardt, L. T. (1991). Psychoneuroimmunology: a literature review. American Journal of Orthopsychiatry, 61(1), 35-47, January.

Wolff, S., Rodin, B., & Cleaver, J. E. (1977). Sister chromatid exchanges induced by mutagenic carcinogens in normal and xeroderma pigmentosum cells. Nature, 265(5592), 347-349, January 27.