INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Human T lymphotropic virus type I (HTLV-I), a human retrovirus endemic in southwestern Japan and other isolated regions of the world, is etiologically associated with adult T cell leukemia (ATL) (1, 2) and several chronic inflammatory diseases such as HTLV-I-associated myelopathy/tropical spastic paraparesis (HAM/TSP) (3, 4), HTLV-I-associated arthropathy (HAAP) (5), and HTLV-I-associated uveitis (HAU) (6). HAM/TSP and HAU are also known to be associated with bronchopulmonary disorders (7). For example, T cell alveolitis was observed in these patients as evidenced by increased numbers of activated T cells expressing interleukin 2 (IL-2) receptor in the bronchoalveolar lavage fluid (BALF) and infiltration of T cells in the alveolar septa. Maruyama and coworkers (14) also reported a strong association between bronchoalveolar disorders and patients with HTLV-I infection and even healthy carriers of HTLV-I and proposed the term HTLV-I-associated bronchopneumonopathy (HAB) for these pathogenic conditions. More recently, the frequency of HTLV-I- infected T cells and level of viral activation were demonstrated to be high in the BALF of these patients (15). These results strongly suggest a possible relationship between HTLV-I and bronchoalveolar disorders. However, the high number of activated T cells in BALF and infiltration of T cells into lung tissues do not necessarily indicate that these changes are due to infection with HTLV-I, because the lung is continuously exposed to a variety of stimulants in inhaled air that may result in the appearance, followed by the disappearance, of inflammatory changes. Thus, the disease entity of bronchoalveolar disorders associated with HTLV-I remains unsubstantiated, in contrast to HAM/TSP (3, 18).

Iwakura and co-workers (19) succeeded in establishing a transgenic mouse that expresses gene segments of HTLV-I, including the env and pX regions. They also demonstrated that these mice exhibited clinical and histopathological manifestations of arthritis, indicating that these mice are an appropriate model for HAAP. Interestingly, the expression of pX mRNA was detected in the lung as well as joint tissues of these mice (19).

In the present study, we analyzed the histopathological changes in the lung tissues of these transgenic mice to examine whether bronchoalveolar disorders occur in these mice. Our results showed inflammatory changes with bronchoalveolar lymphocytosis. These changes were similar to the histopathological findings observed in HTLV-I-infected patients (12, 13). Furthermore, we analyzed the local expression of p40 ^tax mRNA in the lungs and correlated its level to the severity of lung lesions in order to ascertain the role of this viral component in the pathogenic mechanisms.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals

Transgenic mice used in this study were originally established by Iwakura and colleagues (19), who introduced the pX and env regions of the HTLV-I genome with their own long terminal repeat promoter into fertilized mouse ova (C3H/HeN). The original transgenic mice were back-crossed with BALB/c mice. Male or female mice, of generation 10 or 11 after back-crossing, were used in the present experiments. The transgene was detected through dot-blot hybridization using DNA prepared from mouse tails 4 wk after birth (20), and mice with a negative integration of the transgene were recognized as littermates. All mice were housed in a pathogen-free environment and provided with sterilized food and water at the Laboratory Animal Center for Biomedical Science (University of the Ryukyus, Okinawa, Japan). The experimental protocol was approved by the Ethics Review Committee for Animal Experimentation of our university.

Histopathological Examination

Mice were sacrificed at 7, 11, 15, 26 wk of age and the right lower and middle lung lobes were fixed in 4% buffer formalin, dehydrated, and embedded in paraffin. Sections were cut and stained with hematoxylin and eosin, using a standard staining procedure. The sections obtained from littermate and transgenic mice were shuffled and histopathologically examined under a light microscope in a blinded manner.

For quantification of the pathological changes, light microscopic images of the total visual fields in five nonsequential lung sections were captured, digitized, and saved on a Macintosh computer (8100/ 100 AV) using Adobe Photoshop software (version 3.0J). Inflammatory lesions were identified and the area of each selected region was measured using NIH Image analysis software (version 1.61; NIH, Bethesda, MD). The selected areas of inflammatory lesions were measured and summated for each mouse. At the same time, the areas of total fields in the same lung sections were also quantified and summated. The severity of pathological changes in the lungs was calculated by using the following formula: the relative area of lung lesions equals the sum of areas with lung lesions divided by the total areas in five nonsequential sections of the right middle and lower lobes.

Extraction of RNA and Reverse Transcription Polymerase Chain Reaction

Total RNA was extracted from the left lungs of mice 7, 11, 15, and 26 wk of age, as we have described (21). For this purpose, 20 to 45 µg of RNA was obtained from each lung and resuspended in 50 µl of sample RNA solution (15 µl) with 2 µl of hexadeoxyribonucleotide mixture (GIBCO-BRL, Life Technologies, Tokyo, Japan). This solution was incubated for 2 min at 95° C and quickly cooled on ice. In the next step, 12 µl of a solution containing 6 µl of 5 × reverse transcriptase buffer (250 mM Tris-HCl [pH 8.3], 375 mM KCl, 15 mM MgCl₂; GIBCO), 0.5 µl of RNase inhibitor (200 U/ml; GIBCO), 3 µl of 100 mM dithiothreitol, and 2.5 µl of 10 mM dNTP was added, and the tubes were incubated for 2 min at 37° C. We then added 1.0 µl of Moloney murine leukemia virus (Mo-MuLV) reverse transcriptase (RT, 200,000 U/ml; GIBCO) and incubated the sample for 60 min at 37° C. After receiving 45 µl of 0.7 M NaOH and 40 mM EDTA, the tubes were incubated for 10 min at 65° C and quickly cooled on ice. The resultant cDNA was precipitated with 75% ethanol overnight at 70° C. The precipitates were washed once with 75% ethanol, dried, and resuspended in 50 µl of DEPC-treated dH₂O. The samples were stored at 20° C until use. This reaction was always performed simultaneously, using parallel samples in each experiment. The cDNA was then amplified in an automatic DNA thermal cycler (Perkin-Elmer Cetus, Norwalk, CT) using specific primers 5'-ATC CCG TGG AGA CTC CTC AA-3' (sense) and 5'-AAC ACG TAG ACT GGG TAT CC-3' (antisense) for p40^tax (17), and 5'-GTT GGA TAC AGG CCA AGA CTT TGT TG-3' (sense) and 5'-GAT TCA ACT TGC GCT CAT CTT AGG C-3' (antisense) for hypoxanthine phosphoribosyltransferase (HPRT) (21). We added 1.0 µl of the sample cDNA solution to 49 µl of the reaction mixture, which contained the following concentrations: 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 1.5 mM MgCl₂, gelatin (10 µg/ml), dNTP (each at a concentration of 200 µM), 1.0 µM sense and antisense primer, and 1.25 U of AmpliTaq DNA polymerase (Perkin-Elmer Cetus). The preparations in the microtubes were amplified by a three-temperature polymerase chain reaction (PCR) system usually consisting of denaturation at 94° C for 1 min, primer annealing at 55° C for 1 min, and extension at 72° C for 1.5 min. The cycle number was changed every two cycles from 32 to 40 for p40^tax and from 26 to 34 for HPRT.

The PCR process was performed for both genes at the same time for each mouse, and the PCR products were electrophoresed on 2% agarose gels, stained with ethidium bromide (0.5 µg/ml), and examined with a UV transilluminator. The obtained bands of amplified DNA were quantitated using an NIH Image analysis software application, and each value was plotted with the corresponding cycle number. Using this graph, we calculated the differentials in the PCR cycle number that provided the same intensity of amplified DNA between the two genes. The level of expression of p40 ^tax mRNA was expressed as a reciprocal of the value relative to that of HPRT mRNA.

Statistical Analysis

Linear least-squares regression analysis was used to estimate the relationship between p40 ^tax mRNA expression and the severity of pathogenic changes in the lungs. A p value less than 0.05 was considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Histopathological Changes in the Lung Interstitium

Mice were sacrificed at 7, 11, 15, and 26 wk of age, and paraffin sections of the lung were examined histopathologically for the appearance of inflammatory lesions. No pathological changes were observed at any time interval in littermate mice and 7-wk-old transgenic mice. In contrast, in transgenic mice older than 11 wk, there was a significant level of infiltration of inflammatory cells (consisting mostly of lymphocytes) in the interstitial areas of the lung, including peribronchial and perivascular areas and interalveolar septa. The typical views are indicated in Figure 1. The frequency with which mice showed such lesions in their lungs increased with age, although a time-dependent increase in severity was not clearly observed, as indicated in Table 1.

View larger version (117K): [in this window] [in a new window]   View larger version (139K): [in this window] [in a new window]   Figure 1.   Histopathological examination. Littermate (A) and p40tax transgenic mice (B) were sacrificed at 26 wk of age, and the right lower and middle lung lobes were fixed, dehydrated, and embedded in paraffin. Sections were cut and stained with hematoxylin and eosin, and examined (original magnification, ×33) under a light microscope. Each photograph represents a typical view of the entire sections examined in the littermate (n = 6) and transgenic mice with lung lesions (n = 6).

Activation of Transgene in Lung Tissues

To elucidate whether the transgene was actually activated in the lungs of p40^tax transgenic mice, we next examined the expression of p40^tax mRNA using the RT-PCR method. In the lungs of littermate mice, the expression of p40^tax mRNA was not detected throughout the observation period, although HPRT mRNA was expressed at almost a constant level throughout the same period. In sharp contrast, p40^tax mRNA was clearly expressed in the lungs of 7-, 11-, 15-, and 26-wk-old transgenic mice (Figure 2A). Although the expression of p40^tax mRNA was detected in transgenic mice, its level varied from one animal to another. For example, the level of expression differed in three individual transgenic mice, 7 and 11 wk of age, being almost undetected or weak in one animal compared with the other two 7- and 11-wk-old mice (Figure 2A). More interestingly, the level of expression of p40^tax mRNA tended to correlate with the severity of pathogenic changes in the lungs. The level of bronchoalveolar lymphocytosis also correlated with the intensity of p40^tax mRNA expression, with a higher number of lymphocytes occurring with high levels of p40^tax mRNA expression (Figure 2B).

View larger version (53K): [in this window] [in a new window]   View larger version (80K): [in this window] [in a new window]   Figure 2.   Expression of p40tax mRNA. (A) Littermate and p40tax transgenic mice were sacrificed at 7, 11, 15, and 26 wk of age, and total RNA was extracted from the left lungs. Subsequently, RT-PCR was conducted for p40tax. HPRT was used as an internal control. (B) Comparison of the histopathological findings (original magnification, ×13) and PCR results for three 11-wk-old p40tax transgenic mice. Arrows indicate lung lesions with lymphocytic accumulation.

Relationship between Expression of p40^tax mRNA and Lymphocytic Inflammatory Changes

A further study was performed to confirm the relationship between p40^tax mRNA expression and lymphocytic inflammatory changes in the lungs. For this purpose, a semiquantitative PCR was used, in which the level of p40^tax mRNA expression was calculated relative to that of HPRT. Against samples of DNA reversely transcribed from complementary RNA extracted from the lungs of nine transgenic mice, PCR was performed using primer sets for both p40^tax and HPRT at 32, 34, 36, 38, and 40 cycles and at 26, 28, 30, 32 and 34 cycles, respectively. The obtained bands of amplified DNA were quantitated using NIH Image software, and each value was plotted against the corresponding cycle number. As shown in Figure 3A, a good linear correlation was present for each sample. Using this graph, we calculated the relative level of expression of p40^tax mRNA, which was expressed as a reciprocal value relative to that of HPRT mRNA. On the other hand, the light microscopic images of the visual fields in five nonsequential lung sections of the upper and middle lobes were digitized and then analyzed, using the NIH Image software. In this analysis, areas with lymphocytic accumulation were identified, measured, and summated in each mouse. To provide an index for the level of inflammation, the inflammatory changes were expressed as the areas with lymphocytic accumulation (lung lesions) relative to the total areas of five selected lung sections (Figure 3B). In the next step, we plotted the values of the relative severity of pathological changes in the lungs of nine p40^tax transgenic mice against those of the relative expression of p40^tax mRNA. As shown in Figure 3C, there was a significant positive relationship between inflammatory changes and expression of p40^tax in transgenic mice (r² = 0.631, p < 0.05).

View larger version (22K): [in this window] [in a new window]   View larger version (60K): [in this window] [in a new window]   View larger version (17K): [in this window] [in a new window]   Figure 3.   Relationship between p40tax mRNA expression and lymphocytic inflammatory changes in the lung. (A) Against samples of DNA reverse transcribed from complementary RNA extracted from the left lungs of nine transgenic mice, PCR was performed using primer sets for both p40tax and HPRT at 26, 28, 30, 32, and 34 cycles and 32, 34, 36, 38, and 40 cycles, respectively. The obtained bands of amplified DNA were quantitated, and each value was plotted with the corresponding cycle number. Using this graph, we calculated the level of expression of p40tax mRNA, which was expressed as the reciprocal value relative to that of HPRT mRNA. (B) Light microscopic images of the total visual fields in five nonsequential lung sections of the upper and middle lobes were examined to determine areas with lymphocytic accumulation. These images were then digitized and analyzed using NIH Image software. In this analysis, areas with lymphocytic accumulation were measured and summated within each lung field. This process was repeated in each mouse, and the level of inflammatory changes was expressed as the area relative to the total area of the lung section. (C ) Relationship between the relative p40tax mRNA expression and severity of pathological changes.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

On the basis of clinical and epidemiological studies, HTLV-I is considered to be closely associated with certain bronchopulmonary disorders, designated as HTLV-I-associated bronchopneumonopathy (HAB) (7). However, to our knowledge, no direct relationship between viral infection and pathogenic conditions has been established until now. It is possible that the continuous exposure of the lung to a variety of stimulants results in cyclical changes in the histological features, including frequent appearance and disappearance of inflammatory changes. These changes may preclude an accurate diagnosis of HAB, and, thus, this clinical entity remains unsubstantiated since it was first described. To circumvent such difficulties, we used transgenic mice in the present study. These mice express gene segments of HTLV-I, including the env and pX regions. In these mice, we demonstrated the presence of a clear inflammatory process associated with infiltration of lymphocytes in the peribronchial and perivascular areas and in alveolar septa, which resembled the histopathological findings in patients with HTLV-I infection (12, 13). This process did not exist in the first 7 wk, but developed later in life. More importantly, these pathological changes correlated well with the level of local expression of p40^tax mRNA in lungs (Figure 3C). Thus, our study provided for the first time direct evidence showing a relationship between HTLV-I infection and development of bronchopulmonary disorders.

Sugimoto and co-workers (7) demonstrated in HTLV-I-infected patients the presence of lymphocytic inflammatory lesions, which were observed with considerable frequency in the lungs, as indicated by the presence of alveolitis and peribronchiolitis associated with lymphocytic infiltration and a relative increase in activated T cells in BALF. In the transgenic mice, lymphocytic infiltration was evident both in peribronchial/ perivascular areas and alveolar septa. We detected similar histopathological findings in two patients infected with HTLV-I; the first was pathologically diagnosed as having chronic bronchiolitis whereas the other patient was diagnosed as having lymphocytic interstitial pneumonitis (data not shown).

Although the precise mechanism for accumulation of lymphocytes in the indicated areas remains to be elucidated, one possible explanation could involve differences in the distribution of p40^tax mRNA expression, i.e., lymphocytes may accumulate preferentially in areas where the p40^tax gene is activated. Consistent with this hypothesis, a significant relationship was observed between the level of expression of this gene and the extent of lymphocytic inflammatory changes in the lungs. Preliminary results from our laboratory have shown a sufficient level of expression of p40^tax mRNA both in T cells and other leukocyte fractions prepared from lung homogenates of these mice after treatment with collagenase and DNase. Further studies to elucidate the relationship between the distribution of different cell populations expressing the p40^tax gene and inflammatory lesions, using an in situ hybridization method, are currently underway in our laboratory.

Tax is known to act as a trans-activation factor not only on the long terminal repeat of HTLV-I itself but also on the promoters of many cellular genes including cytokines, cellular oncogenes, and cell surface molecules (1). Iwakura and colleagues (22) indicated that genes for several inflammatory cytokines, such as IL-1, IL-1, IL-6, TNF-, TGF-1, IFN-, and IL-2, were activated in the inflammatory joints of p40^tax transgenic mice. In addition, an interesting finding has been reported by Baba and colleagues (23), who demonstrated that both HTLV-I-infected and p40^tax-transfected cells produced a large amount of chemokines including IL-8, IP-10, MIP-1, and MIP-1. In preliminary studies from our laboratory, both inflammatory cytokines and chemokines were detected at the mRNA level by RT-PCR, and their expression seemed to correlate with that of p40^tax mRNA in the lungs of p40^tax transgenic mice. Although direct evidence has not yet been obtained, these findings may suggest that chemokines, produced by HTLV-I-infected cells (23) or indirectly by uninfected cells on stimulation with proinflammatory cytokines derived from infected cells (1, 22, 24), are involved in the pathogenesis of HTLV-I-associated lung disorders.

Our results showed that in the p40^tax transgenic mice, lymphocytic inflammatory lesions developed in peribronchial/ perivascular areas and alveolar septa after birth and progressively increased in incidence later in life, as reported in the joint lesions (19), although a time-dependent increase in severity was not clearly observed. The severity of the histopathological changes correlated significantly with the level of expression of p40^tax mRNA in the lungs. These findings indicate that the expression of p40^tax is directly involved with lymphocytic infiltration and induced the inflammatory lesions in lungs. Considered collectively, our results indicate that the transgenic mice used in our study represent an appropriate animal model for the elucidation of pathogenic mechanisms of HTLV-I-associated pulmonary disorders. On the other hand, HTLV-I-associated lung diseases are usually observed in patients with HAM/TSP (7, 12, 13, 15, 16), whereas these mice did not show any neurological symptoms in spite of the expression of p40^tax mRNA in brain tissues (Reference 19; and our unpublished observation, 1998). Therefore, it also should be noted that the present animal model may not completely represent the pathogenesis of HTLV-I-associated lung diseases complicated in patients with HAM/TSP.