Histological Investigation of Diagnostically Challenging Tubule Profiles in Advanced Chronic Progressive Nephropathy (CPN) in the Fischer 344 RaT

Abstract

Recently, guidelines were suggested for discriminating proliferative-appearing tubule profiles encountered in advanced spontaneous chronic progressive nephropathy (CPN) of rats, from hyperplastic precursors of renal tubule tumors (Hard and Seely, 2005). These recommendations were based on histological evaluation of a large number of cases of severe to end-stage CPN in male F344 rats from 8 separate 2-year carcinogenicity studies held in the Archives of the National Toxicology Program, NIEHS. This work has now been extended to characterize the various lesions further, mainly by serial sectioning to track their origin and fate within the adjacent renal tissue, but also by applying special staining procedures such as immunohistochemical assessment of proliferative activity, as well as fluorescence microscopy, to seek further differences from atypical tubule hyperplasia. The results obtained from these additional investigations support the contention that certain tubule profiles with a misleading proliferative appearance, sometimes found in advanced CPN, should be distinguished from preneoplastic tubule foci, and regarded as components of the nephropathy process.

Introduction

In routine 2-year carcinogenicity studies using conventional strains of rat, the frequency and severity of chronic progressive nephropathy (CPN), an age-related disease of spontaneous origin, can become a confounding factor in the interpretation and diagnosis of proliferative tubule profiles and their relationship to neoplastic development (Hard and Khan, 2004). This problem arises because proliferative-appearing or regenerative tubule profiles are relatively common in advanced to end-stage CPN, and difficult to distinguish from tubule preneoplasia. Furthermore, certain chemicals, including a number tested in 2-year studies by the National Toxicology Program (NTP), National Institute of Environmental Health Sciences, have enhanced the severity of CPN to an advanced state (Eustis et al., 1994; Lock and Hard, 2004), often resulting in an end-stage kidney, as was the case with hydroquinone (Hard et al., 1997) and ethyl benzene (Hard, 2002).

To ensure that the hazard assessments and classification of chemicals, particularly those that have caused CPN exacerbation, are not compromised, it is important to be able to discriminate CPN-related tubules of little biological significance from genuine preneoplastic hyperplasia. There is general agreement that atypical tubule proliferation, termed atypical tubule hyperplasia (ATH) in some classifications (Alden et al., 1992; Hard et al., 1995), is an obligatory precursor of renal tubule tumor development in the rat (Hard, 1986; Lipsky and Trump, 1988; Dietrich and Swenberg, 1991; Nogueira et al., 1993). This view has culminated over the years from the consistent results of many studies on renal tumor pathogenesis using a wide variety of chemical carcinogens, including aflatoxin B1 (Epstein et al., 1969), dimethylnitrosamine (Hard and Butler, 1971), N-ethyl-N-hydroxy-ethyl nitrosamine (Hiasa et al., 1979), N-(4′-fluoro-4-biphenyl)acetamide (Dees et al., 1980), N-nitrosomorpholine (Bannasch, 1984), and ochratoxin A (Boorman et al., 1992). Criteria for diagnosing ATH have been published by the WHO International Agency for Research on Cancer (Alden et al., 1992) and the Society of Toxicologic Pathology (Hard et al., 1995).

Recently, we conducted a histopathologic survey of a large sample of rats with advanced and end-stage CPN, identifying unusual tubule profiles that were integral components of CPN, and providing recommendations for their discrimination from ATH (Hard and Seely, 2005). In that survey, we accessed 2-year carcinogenicity studies for 8 chemicals held in the NTP Archives, concentrating on hematoxylin and eosin (H&E) stained male rat kidney sections. We have now carried out additional histological procedures, including serial sectioning and some special staining methods, in order to further characterize the nature of the unusual CPN-associated profiles and to highlight differences from ATH. The new results reported here are intended to be used in conjunction with, and to complement, the findings described previously in the Hard and Seely (2005) report.

Materials and Methods

The 8 chemicals in the NTP carcinogenicity testing program evaluated for CPN severity and renal tubule tumors and neoplastic precursors in male rats (Hard and Seely, 2005), were: quercetin (NTP, 1992), coumarin (NTP, 1993), ozone (NTP, 1994), oxazepam (NTP, 1998a), chloroprene (NTP, 1998b), tetrahydrofuran (NTP, 1998c), isobutene (NTP, 1998d), and primidone (NTP, 2000). From these studies, specimens representing various morphological forms of proliferative tubule profiles were selected for further histological processing and characterization, as described next.

Severity Grading for CPN

The severity of CPN was graded semiquantitatively on a scale of 0–8 as described by Hard and Khan (2004). In this schema, grades 1–5 (minimal to high-moderate) represent a progressive increase in the number of focal lesions of CPN; grade 6 (low-severe), as the point where foci begin to coalesce into areas of cortical change; grade 7 (high-severe), where a majority of the outer parenchyma is affected by CPN; and grade 8 (end-stage), where no or almost no normal parenchyma remains. This final stage signals imminent death from chronic renal failure. In this study, advanced CPN was considered to span the grades from 6 through 8, but specimens were selected mainly from rats with high severe or end-stage disease (grades 7 and 8).

Special Stains

Selected kidney sections with specific CPN-related tubule lesions were stained with Masson’s trichrome (Sigma-Aldrich Accustain, St Louis, MO), periodic acid Schiff reagent (PAS) (AFIP, 1992), Lillie’s allochrome for discriminating basement membrane and collagen (AFIP, 1968), and Luna’s stain for differentiating new from mature collagen (Luna, 1992).

Serial Sectioning

The paraffin blocks of left and right kidneys from four male rats with end-stage (grade 8) CPN from the chloroprene and primidone 2-year studies were serially sectioned at 5 μ-intervals for a total of 50 sections. Each section in successive ribbons was picked up individually and placed on a sequentially numbered glass slide. The serial sections were stained with H&E, with every 10th section being immunohistochemically processed for anti-PCNA.

Immunohistochemical Staining for DNA Synthesis

DNA synthesis, as an indicator of cell proliferation, was demonstrated on paraffin-embedded sections by the immunohistochemical method for proliferating cell nuclear antigen (PCNA), employing the ZymedPCNA staining kit (ZYMED Laboratories Inc., San Francisco, CA). This method uses a biotyinylated PCNA monoclonal antibody, clone PC 10 (60 minutes), with streptavidin-peroxidase as a signal generator (10 minutes), and dimethylaminobenzidine as the chromogen (5 minutes). After the deparaffinization and hydration procedures, the samples were subjected to an antigen retrieval step by microwave before further processing (2 individual 5-minute intervals at 50% power). Hydrogen peroxide (3%) was used to block endogenous peroxidase (10 minutes). The slides were counterstained in Mayer’s hematoxylin.

For comparison of proliferative rates, PCNA-labelled images of 3 samples of solid CPN tubule profiles and 2 samples of ATH were quantified for cells undergoing DNA synthesis using Image Pro Plus software (Media Cybernetics, Silver Spring, MD).

Fluorescence Microscopy

Multiple specimens of H&E-stained CPN tubule profiles, foci of ATH, and adenomas, were examined for autofluorescence at a wavelength of 450 nm using an Olympus AX80 microscope.

Results

In Hard and Seely (2005), the morphological forms of tubule profile associated with advanced to end-stage CPN were described as “intratubular cell clusters,” “dilated tubule profiles,” and “solid tubule profiles.” These same descriptions are used here for convenience, but as stated previously (Hard and Seely, 2005), we are not recommending the adoption of these terms for general use. We advocate the inclusion of such lesions under the generic term CPN tubule profile, or simply acknowledged as part of the spectrum of CPN pathomorphology.

Special Stains

With the various connective tissue stains applied to kidneys affected by severe and end-stage CPN, including Masson’s trichrome, PAS, Lillie’s allochrome, and Luna’s stain, it was confirmed that the “solid tubule profiles,” and most “dilated tubule profiles,” were characterized by a surrounding band of very thickened basement membrane, which merged into expanding collagen. Lillie’s allochrome in particular, was useful for discriminating the thickened tubular basement membrane, which stained red to purple, from the surrounding, blue-green staining collagen (Figure 1). Both Lillie’s allochrome (Figure 2) and PAS staining also highlighted the tortuosity of the thickened basement membrane surrounding the many atrophic tubules present. With Luna’s stain, it was observed that the majority of collagen occupying the intertubular space in severe and end-stage CPN represented an immature (blue-staining) rather than a mature (red-staining) form of collagen (Figure 3). Many of the CPN-affected tubules appeared to be partially or completely collapsed because of the increasing fibrotic process, in keeping with the implication that CPN might be a disease of expanding extracellular matrix (Abrass, 2000). A previous immunohistochemical study of the interstitial fibrosis in CPN identified abnormal deposition of collagen types I, III, and IV, along with fibronectin and tenascin in fibrotic areas adjacent to dilated or atrophic tubules (Nakatsuji et al., 1998).

Serial Sections

Serial sectioning at 5 μshowed that “intratubular cell clusters” were very small entities, appearing and disappearing within a short distance, usually less than 30 μ(Figures 4a–d). They appeared to represent a clustering of small basophilic cells at bends or infoldings in tubules that were at the most recent stage of involvement in the CPN process. Serial sectioning also demonstrated that these profiles were associated with a central core of basement membrane that was continuous with the peritubular membrane, indicative of an indentation or bend in the tubule.

Solitary “dilated tubule profiles” had a very expanded lumen, were surrounded by overtly thickened basement membrane, and sometimes showed the appearance of a low, polypoid proliferation protruding from the cuboidal lining. Serial sectioning demonstrated that these dilated tubule profiles could persist with no significant morphological change over a substantial distance except that the minor complex proliferations or protruberances of the lining epithelium were discontinuous along the tubule length. However, “dilated tubule profiles” were observed to originate and eventually terminate in atrophic tubules (Figure 5a–d).

“Solid tubule profiles” were invariably located within a large island of very thickened basement membrane and expanded matrix, in which similar but smaller tubule profiles were also embedded. Serial sectioning showed that the multiplicity of profiles in these islands represented convolutions of a single CPN-affected tubule (Figure 6a–d).Tracing the fate of a single large solid profile through consecutive serial sections demonstrated that, like the “dilated tubule profiles,” they appeared to originate and terminate in atrophic tubules (Figure 7a–d), but did so over a shorter distance.

In the series of kidneys in which serial sectioning was performed, only one focus of ATH was found. However, from the many foci of ATH observed in our survey of the various NTP carcinogenicity studies, it was noted that ATH sometimes persisted through step sections, which had been cut at 40–50μ intervals. They appeared to originate in tubules that, apart from basophilia, had retained their normal size and appearance, and were not associated with atrophic tubules. The one adenoma present in the samples of serial sections, an incipient lesion, persisted through all 50 5 μ-sections.

PCNA Immunohistochemistry

In severe and end-stage kidney (grades 6–8), 2 tubular components of the usual CPN spectrum showed a relatively high degree of proliferative activity. These were the well-formed basophilic tubules that appeared to have been most recently incorporated into the CPN process, and series of nondistinctive dilated tubules lined by a single layer of cuboidal epithelium (Figure 8). Tubules containing hyaline proteinaceous casts (lined by very flattened epithelium), and atrophic tubules, were negative for anti-PCNA labeling (Figure 8). In addition, among the distinctive tubule profiles described in the Hard and Seely (2005) report, “intratubular cell clusters” (Figure 9), and hypertrophic tubules, lacked PCNA reactivity. “Solid tubule profiles” set in an area of expanded basement membrane, including those of relatively large size, showed only a low level of positive staining (Figure 10). In striking contrast, foci of ATH and incipient adenomas were hotspots for anti-PCNA labeling, reflecting a high proliferative rate (Figure 11). Quantification of the percentage of cells in DNA synthesis for 2 samples of ATH showed a range of 14.5–16.4%, compared to 0–1.2% for 3 samples of “solid tubule profiles.”

Fluorescence Microscopy

When viewed under fluorescent illumination, “intratubular cell clusters,” “dilated tubule profiles,” and “solid tubule profiles” showed no autofluorescence internally within the cell mass or lining. In contrast, foci of ATH (and adenomas) showed bright autofluorescence due to patches or wisps of basement membrane insinuating between cell groups within the profiles (Figure 12a, b).

Discussion

This histological investigation of tubule profiles of unusual appearance encountered in severe and end-stage CPN, using special stains, serial sectioning, immunohistochemistry for DNA synthesis, and fluorescence microscopy, has underscored the biological difference between this spectrum of CPN lesions and those representing the precursor stage of renal tubule neoplasia. The data add confirmatory support to our recommendations (Hard and Seely, 2005) that such lesions should not be identified as preneoplastic in 2-year carcinogenicity bioassays where advanced CPN is a confounding factor.

In particular, serial sectioning was informative in characterizing the origin and fate of intratubular cell clusters, dilated tubule profiles, and solid tubule profiles. The “intratubular cell clusters” were confined to proximal convoluted tubules that were undergoing early incorporation into the CPN disease process, and were very small lesions, negative for proliferative activity. Solitary “dilated tubule profiles” and “solid tubule profiles” appeared to be connected at each end with atrophic tubules, but in the intervening distance, showed none of the features of ATH described in our previous report (Hard and Seely, 2005).

In this respect, it was significant that they lacked any margination of fibroblasts throughout their length, which would have been indicative of an expanding lesion (Hard and Seely, 2005). Solitary “dilated tubule profiles” sometimes appeared to have an increased number of epithelial lining cells in the form of low polypoid protruberances, but these were discontinuous along the length of the tubule dilation. As Risdon and Woolf (1998) have pointed out, cellular proliferation would be expected in the lining epithelium of some dilating tubules in order to accommodate the expanding luminal volume. The alternative would be a stretching and flattening of the lining cells, as is also seen in CPN when tubules become dilated with hyaline casts or colloid, and accordingly, these latter proved to be negative for proliferative activity.

Furthermore, “solid tubule profiles,” which were sometimes the most difficult CPN lesion to distinguish from preneoplastic foci, were typified by a level of cell turnover that was demonstrably less than was characteristic of ATH and small adenomas. In addition they lacked the autofluorescing wisps or patches of basement membrane that were present in all of the ATH and adenomas examined under fluorescent illumination. As suggested previously (Hard et al., 1997; Hard, 2002), “solid tubule profiles” might represent an attempt by the failing kidney to regenerate tubule epithelium, particularly as these profiles involved tubules that were otherwise undergoing atrophy.

In conclusion, this histological investigation provides further support for the contention that the diagnostically challenging tubule profiles often occurring in advanced cases of rat CPN, particularly where there has been exacerbation of CPN severity by the test chemical, are biologically different from preneoplastic tubule lesions. Significantly increased PCNA labeling for DNA synthesis, and delicate basement membrane extensions into the cell mass, add to the criteria identified in Hard and Seely (2005) for discriminating ATH from CPN tubule profiles. The finding that “dilated” and “solid” tubule profiles represent modifications of atrophic tubules also reinforces the recommendation that such profiles should not be reported as separate entities from CPN in carcinogenicity studies.