Clinical and Biochemical Characteristics

Differentiating Chronic Fatigue Syndrome

from Major Depression and Healthy Control Populations:

Relation to Dysfunction and RNase L Pathway

Abstract:

Patterns of immune dysfunction have emerged in chronic fatigue syndrome (CFS) that include an immune activation state (evidenced by increased activated T lymphocytes and circulating cytokines) and poor cellular function (low natural killer (NK) cell cytotoxicity and impaired T lymphocyte response to mitogens).

Therefore, the aim of the current study was to examine the relationship between clinical and functional characteristics, immune abnormalities and sta­tus of the RNase L pathway in CFS compared with healthy control and depression control populations.

All study participants were assessed with respect to their general health, functional status, blood count and chemistry, biochemical and immune pa­rameters. The CFS group (N = 66) demonstrated clinical, functional and biochemical abnormalities distinct from the healthy (N = 62) and depression (n = 51) control groups. The CFS group showed marked functional impair­ment compared with both control groups (P < .001) as measured by the Medical Outcomes Study 36-Item Short Form Health Survey (SF-36) (P < .001).

The CFS group also showed decreased cognitive performance on a computerized test battery compared to healthy (P < .001) and depression controls (P < .009) and significantly higher 37/80 kDa RNase L ratio (P < .001) compared with both control groups.

The odds ratios of a 37/80 kDa RNase L ratio > 2 compared with the CFS patients were 3.9 for the healthy controls (95% confidence limit (CL) 1.0-15.2, P < .05) and 65.8 for the depression controls (95% CL 10.7-406.6, P < .001).

The CFS group demonstrated low NK cell cytotoxicity compared to healthy controls (P = .045).

The correlation between abnormalities in the RNase L pathway and impaired NK-cell function (r = .21, P < .006) suggests that both may be part of the sa­me underlying disease mechanism, at least in this homogeneous population of very disabled CFS patients.

Healthy contact-control subjects who had exposure to CFS patients showed a number of characteristics similar to the CFS patients, including an increased mean 37/80 kDa RNase L ratio (P < .04) and prevalence of the 37/80 kDa RNase L ratio > 2 (P < .03).

In these contact-control subjects, the 37/80 kDa RNase L ratio was correlated with the interferon-α levels (r = .58, P < .02), suggestive of activation of the interferon pathway.

The results of the present study support the cytokine/immune activation model in this well-characterized CFS patient group.

DISCUSSION

An accumulating body of evidence indicates that CFS is characterized by a state of chronic immune activation. Many CFS patients exhibit immune dysfunction, including increased numbers of activated lymphocytes (i.e. activated CD8+ cytotoxic T cells), decreased NK cell number and function, elevated immune complexes, and elevated levels of circulating cytokines (7-15,37-39).

These changes are potential evidence of a humoral response to an infectious agent, although no etiologic agent(s) for CFS has been conclusively identified. The immune activation state observed in CFS could possibly explain many of the signs and symptoms of the disorder, as many different cytokines are associated with symptom development in humans (14, 40-43).

Our working hypothesis has been that dysfunction of the dsRNA-dependent, interferon-inducible RNase L pathway is involved in the immune activation ob­served in CFS. This pathway is part of the innate antiviral defense mechanism of mammalian cells (17,18). The pathway also regulates cell growth and differentiation (17,18).

Our initial studies of individuals with CFS revealed an elevation in basal levels of the RNase L pathway (19,20). In subsequent studies, we discovered a novel low molecular weight 2-5A-dependent 37 kDa form of RNase L in CFS (21). We’ve previously reported a significant correlation between the deregulation of the RNase L pathway and the disability of individuals with CFS, i.e. deregulation of the pathway is an indication of a lower state of general health (44).

A highly significant negative correlation was also observed between the native 80 kDa RNase L and the 37 kDa RNase L, suggesting that the 37 kDa RNase L might be derived from the 80 kDa RNase L (44). We postulated that the 37 kDa RNase L isoform might be the results of proteolytic degradation of the 80 kDa RNase L (23). Increased proteolytic activity in PBMC extracts from CFS patients has been shown to result in accumulation of a low molecular weight form of RNase L with maintenance of 2-5A dependent nucleolytic activity (27).

A number of independent investigations have observed deregulation in the RNase L pathway in CFS (24-29). In a study in which CFS patients were compared with control subjects, the presence of the 37 kDa Original Research 27 isoform of RNase L in PBMC extracts distinguished CFS from fibromyalgia or depression (24).

A subsequent study reported that the 37/80 kDa RNase L ratio discriminated between CFS patients and controls with high sensitivity and specificity (28).

Elevated 37/80 kDa RNase L ratio in CFS has also been shown to be associated with a decrease in VO2 max and lower exercise capacity (26). A decrease in RNase L inhibitor (RLI) mRNA expression and an increase in total RNase L enzyme activity in CFS compared with controls have been reported (29).

Another study reported no difference in RNase L or RLI mRNA expression in CFS, but unfortunately did not assess protein activity (45). The impact of the 37 kDa RNase L was not addressed in either study. The lack of clinical or functional assessment of the participants in either study makes it impossible to assess the CFS patient groups.

In addition to a healthy control group, a depression control group was inclu­ded in this study. Individuals with depression and CFS share many symptoms, including fatigue. Up to 80% of CFS patients report concurrent depression or anxiety (33,46-50). A strong association between chronic fatigue and psychological distress has been found, but no evidence for genetic covariation (51). In addition, depression is a potentially confounding factor in the measurement of immune function.

To our knowledge, the present study is the first to report a detailed examination of the relationship between the clinical and functional characteristics, immune abnormalities and the status of the RNase L pathway in CFS.

The three study groups described here were very distinct, as demonstrated by multivariate analysis of all clinical, immunological and biochemical data (Figure 4). The functional status of the three study groups was consistent with previous reports; the CFS study subjects showed marked functional impairment compared with both the healthy and depression control subjects (Table 3, Figure 1).

Our results are in excellent agreement with previous studies in the United States and Europe, in which the severity and pattern of functional impairment as measured by the SF-36 distinguishes between CFS study subjects and depression (33,49,52). As expected, the depression controls, due to their chronic medical condition, have a quality of life or functional state of well being that is somewhat impaired compared to healthy controls.

However, they have an overall functional status much higher than CFS pa­tients, which underscores the severity of functional impairment experienced by this cohort of CFS patients.

We have confirmed the historical data on an abnormality in NK cell function in CFS (Figure 3). Our results are consistent with the decreased NK cell cytotoxicity observed in CFS compared to healthy controls (8,9,12,15).

In addition, the NK cell dysfunction in CFS was correlated with elevated 37/ 80 kDa RNase L ratio in the entire study population (r = .21, P < .006) (Table 5, LU20/CD56). This correlation trend between the abnormalities in the RNase L pathway and the impaired NK cell function suggests that they may be a part of the same pathogenetic mechanism demonstrated in the disorder.

Impaired NK cell function may be consistent with the report of decreased nitric oxide-mediated NK cell activation in CFS (16). A role for nitric oxide synthase in the etiology of CFS has been proposed (53) but further studies are required to examine these possibilities.

Correlation analysis of the 37/80 kDa RNase L ratio revealed some very interesting associations in the entire study population.

The CFS group exhibited more cognitive dysfunction than either the healthy control or depression control groups (P < .001) (Table 4). Additionally, the 37/80 kDa RNase L ratio was weakly correlated with cognitive dysfunction within the CFS group as measured by the ACPT (Table 5). Although a vast majority of CFS patients report cognitive problems, cognitive dysfunction in CFS has been difficult to document (54-58). Our results are consistent with the literature reporting that the cognitive impairment observed in CFS cannot be explained by coexisting depression (58). Further assessment is required using the individual components of the ratio to better understand the relationships.

Although essential as part of the clinical work-up of a patient to eliminate other possible causes of fatigue, the physical examination findings and blood chemistry screening studies were mostly unremarkable. The intergroup ANOVA assessment found few changes in mean blood or blood chemistry scores (Table 3).

However, the correlation analyses (Table 5) revealed that several parameters that were altered in the ANOVA analysis (i.e., the ESR, alkaline phosphatase and chloride levels) were also correlated with the 37/80 kDa RNase L ratio. These changes will be assessed in a subsequent paper. However, the in­crea­sed ESR, alkaline phosphatase and chloride levels do suggest that the dere­gu­lated RNase L pathway is associated with changes in homeostasis other than alterations in immune and cytokine-mediated events.

Some interesting differences were observed within the control group when contact and non-contact controls were compared. The distinction between contact and non-contact controls has been of interest because CFS is well documented to occur in both epidemic and endemic settings. In contact controls, the post hoc analysis revealed that the 37/80 kDa RNase L ratio,interferon-a, sodium, liver enzymes (AST, ALT) and glucose were elevated, while chloride and the platelet count were decreased (all P values were < .05) (Table 6). In the contact control subjects, the 37/80 kDa RNase L ratio correlated with the IFN-a level and decreased platelet count (Table 7). These correlations were not found in the CFS patients (data not shown).

Thus, The contact controls appear to show evidence of an IFN-a mediated change, which is associated with increases in total RNase L activity and 37/80 kDa RNase L ratio. This suggests that the normal response of the IFN pathway is occurring within the control group and is more pronounced in the contact controls than in the non-contact controls.

It is unlikely that this response is due to a single infectious agent because the controls were enrolled in the study at random over an extended time period. These changes differ from those seen in CFS patients (data not shown) and will be discussed in a separate publication. The differences between CFS patients and the contact controls may be associated with deregulation of the RNase L pathway as has been suggested by several investigators (28-30).

It is noteworthy that another study also observed elevated 37/80 kDa RNase L ratio in contact controls (24).

Type I interferons play a principle role in innate immune response (18). Increases in interferon-a are associated with many alterations in chemistry and pathology, including reduction of platelet count (59), alterations in alkaline phosphatase activity (60,61), reduced basophil counts (62-64), increases in neutrophil counts and activation (64), neutrophil secretory activity (65), alterations in blood urea nitrogen and creatinine (66) and neurological neuritis similar to that seen in multiple sclerosis (67). Many of these changes were observed in this study cohort. Increased neutrophil apoptosis and higher plasma levels of transforming growth factor 1ß have recently been demonstrated in CFS compared to healthy volunteers (68). The authors state that these abnormalities may be indicative of a persistent viral infection or a toxic state.

While these data suggest that IFN-a may have a significant influence upon the functioning of the RNase L pathway within the control subjects, especially in the contact controls, other factors are certainly involved in the development of symptoms in CFS. Examination of these potential relationships will be undertaken in a separate publication. The results of this study should be interpreted with some caution. The group of CFS patients studied here was less heterogeneous than in many previous studies. The inclusion criteria for the current study were designed to select CFS patients who were significantly impaired by their illness. Further study will be required to determine the degree to which the clinical, functional and biochemical abnormalities observed in the CFS study group can be extrapolated to the larger group of patients who meet the criteria for CFS.

In summary, this study examined a number of attributes of clinical presentation, functional status, immune function and the RNase L pathway in a cohort of CFS patients and two well-defined control populations. The results of the present study are consistent with the immune activation model of CFS, but also add the possibility of additional biochemical changes not of obvious immune or cytokine-mediated origin.

Taken together, our results do not support the contention that CFS is simply a form of depression. The abnormalities demonstrated in the RNase L path­way, i.e. increased 37/80 kDa RNase L ratio, can reliably identify a relatively homogeneous subset of patients within the larger group that comply with the working definition of CFS.

Observation of an elevated 37/80 kDa RNase L ratio is a quantitative measure that correlates well with the severity of CFS symptoms and with low NK cell function, at least in this relatively homogeneous population of very disabled CFS patients. This subset of patients demonstrates specific clinical, functional and biochemical abnormalities that are distinct from depression controls and healthy controls.

This observation could potentially provide the clinician with a relatively inexpensive and standardized tool for identifying this subset of patients within the otherwise heterogeneous disease of CFS. Some of the evidence presented indicates that the elevated 37/80 kDa RNase L ratio may also be an indicator of illness severity as measured by the SF-36.

Our results underscore the importance of documentation of clinical and functional parameters using standardized instruments alongside biochemical and immune parameters in research studies on CFS.