Report of Workshop on Lymphangioleiomyomatosis

	INTRODUCTION

TOP INTRODUCTION CLINICAL ASPECTS OF LAM SMOOTH MUSCLE DIVERSITY,... TUMOR-SPECIFIC ANTIGENS AND... RECOMMENDATIONS REFERENCES

Lymphangioleiomyomatosis (LAM) is a rare lung disease affecting primarily young women of childbearing age. The disease is characterized by an abnormal proliferation of smooth-muscle-like cells that over time can grow to obstruct airways, lymphatics, and blood vessels. Little progress has been made in understanding the cause and pathogenesis of LAM since the first description of the disease appeared in the scientific literature in 1937.

The purpose of the workshop was to review what is currently known about LAM and to recommend directions of scientific investigation that will likely contribute to a better understanding of LAM. The workshop addressed three major topics: clinical aspects of LAM; smooth muscle diversity, growth, and signaling pathways; and tumor-specific antigens and regulatory proteins in LAM and tuberous sclerosis complex.

	CLINICAL ASPECTS OF LAM

TOP INTRODUCTION CLINICAL ASPECTS OF LAM SMOOTH MUSCLE DIVERSITY,... TUMOR-SPECIFIC ANTIGENS AND... RECOMMENDATIONS REFERENCES

Information on the clinical aspects of LAM has been gathered in countries throughout the world (1). Studies of LAM in the United Kingdom have supported the hypothesis that disease progression is hormone-dependent; LAM is found exclusively in women, perhaps worsened by pregnancy, and exhibits a slower rate of decline in lung function after menopause. Patients with LAM appeared less likely than a control population to have been pregnant or to have had children, and tended to have had more spontaneous abortions and a greater incidence of fibroids. In the United Kingdom, the mean age of onset was 34 yr. A retrospective study of patients in the United Kingdom showed that the decline in the TL_CO was significantly less in those patients treated with progesterone. Effects of progesterone on decline in FEV₁ were not significant. There was a suggestion that the rate of decline was lower postmenopausally. It appeared that patients with greater impairment of lung function responded better to progesterone.

In Japan, a questionnaire surveyed the occurrence of LAM in 125 hospitals. Sixty-five cases of LAM were reported from 1987 to 1996 in 41 hospitals. Diagnosis was based on CT in 18% of cases, with CT plus pathologic findings used in diagnosis of 53 patients; mean age was 35.7 yr. Subjective symptoms caused by LAM were found in 78% of cases, with dyspnea in 43%, and pneumothorax in 37%. No therapy was shown to improve prognosis; in 11 of 14 deaths, the cause was LAM. The occurrence of LAM per 100,000 population differed between the countries in Asia, with the figure in Singapore being 0.241 and that in Korea being 0.029. The basis for this difference, whether it be real or a reflection of frequency of diagnosis, remains to be determined. Clinical features appeared to be less pronounced, and the survival rate was greater in Europe than in the United States.

In France, cases of LAM were reported as one of a group of rare pulmonary disorders. For inclusion, diagnosis of LAM requires a high resolution CT and one other criterion. It appears that there has been an increase in the diagnosis of LAM recently, possibly because of the use of high resolution CT. The overall prevalence of LAM in France was 1.8 per million women between the ages of 20 and 69, with an annual incidence of 0.3. The mean age of onset was 36.3, and the mean age of diagnosis was 39. The TL_CO was decreased in 82% of patients at diagnosis. Ground glass opacities were observed in 13% on chest radiograph, but these appeared to be transient. Renal angiomyolipomas were found in 29%, with uterine leiomyomas in 37%; 75% of patients were alive 8.5 yr after onset.

LAM should be included in the differential diagnosis of chronic interstitial lung diseases (5). Eosinophilic granuloma represented the major alternative in the differential diagnosis. Eosinophilic granuloma is found predominantly in smokers (> 90%), with a prevalence of 205 cases per million. The disease is characterized by upper and middle lung involvement with multiple thin-walled cysts. Histopathology may be diagnostic and shows a stellate pattern of fibrosis and Langerhans' histiocytes. Electron microscopic examination reveals the presence of Birbeck bodies. The clinical course may be extended and is characterized in most patients by either stabilization or resolution. Eosinophilic granuloma can be distinguished from LAM by its upper lobe predominance and the presence of a nodular component. The Langerhans' histiocyte with surrounding inflammation is found only in eosinophilic granuloma. Idiopathic pulmonary fibrosis can be distinguished from LAM because of its predilection for the peripheral lung fields and the nonuniform distribution of cysts, although sarcoid can present with cysts throughout the lung.

Imaging of the LAM lung has now been performed in a significant number of cases (6). Ventilation scans were characterized by heterogeneous defects and a speckling pattern. The overall grade of disease based on CT did not correlate with cyst size or with speckling. Speckling pattern correlated with cyst size, not CT grade. Univariate analysis showed a correlation with FEV₁. Abdominal imaging commonly revealed adenopathy and the presence of renal masses (46 in 20 patients), consistent with angiomyolipoma.

Histopathologic features of LAM are pathognomonic (7, 8). The disease is characterized by diffuse cystic changes in the lung; the walls of the cysts are lined by smooth muscle, with spaces within the muscle caused by lymphatics. The smooth muscle cells ("LAM cells") are heterogeneous in appearance. Although LAM is not characterized by fibrosis, there is collagen between the muscle cells. Microscopic hemorrhage is common, probably secondary to the penetration of the vessel walls by LAM cells. The LAM cells react with monoclonal antibody HMB45, which was generated against melanoma antigens, and which also reacts with angiomyolipomas, clear cell tumors of the lung, and melanoma cells. Of diagnostic importance, HMB45 does not react with normal smooth muscle cells. LAM cells resemble smooth muscle cells in that they also react with antibodies against actin, desmin, and vimentin.

Lung transplantation is now being performed for LAM and other pulmonary diseases (9, 10). The outcome in LAM is similar to that in COPD, cystic fibrosis, and ₁-antitrypsin deficiency. Open lung biopsy revealed that severe cystic disease, but not smooth muscle proliferation, correlated with a poorer prognosis. Guidelines for transplant include progression despite medical treatment, FEV₁/FVC < 50%, TLC > 130%, FEV₁ < 30%, and severe cystic disease on high resolution CT. LAM recurrence in the transplanted lung was reported in two cases, leading to the proposal that continued treatment with hormonal therapy may be warranted.

To obtain more data on LAM and to promote research on the disease, the National Heart, Lung, and Blood Institute is sponsoring a volunteer national registry for patients with LAM (11). The registry Data and Coordinating Center is located at the Cleveland Clinic Foundation and patients can be enrolled through six major registry clinical centers: Cleveland Clinic Foundation, Mayo Clinic-Rochester, National Heart, Lung, and Blood Institute, National Jewish Medical and Research Center, New England Medical Center, and Stanford University Medical Center or by a patient's personal physician.

	SMOOTH MUSCLE DIVERSITY, GROWTH, AND SIGNALING PATHWAYS

TOP INTRODUCTION CLINICAL ASPECTS OF LAM SMOOTH MUSCLE DIVERSITY,... TUMOR-SPECIFIC ANTIGENS AND... RECOMMENDATIONS REFERENCES

Modulating cell cycle regulatory proteins in restenosis after angioplasty or in paradigms of vein grafting, and, more specifically, the role of cell cycle inhibitor proteins in controlling proliferation of vascular smooth muscle cells has been examined (12). Two families of cell cycle inhibitors were described in various cell types: the kip/cip family (p21, p27, p57) and the ink family (p15, p16, p18, p20). P21 and p27, which have been best studied in vascular smooth muscle, prevent progression from G1 to S phase by inhibiting both the cyclin E/cdk2 and the cyclin D/cdk4 complexes. Whereas the kip/cip family can regulate smooth muscle cell proliferation, a role for ink family proteins has not been established. Studies from several laboratories have shown that p21 and p27 are expressed constitutively at very low levels in systemic arteries but are increased in medial and intimal cells after vascular injury. However, it appears that the increase occurs days to weeks after injury at a time when neointima has already developed. Using adenoviral vectors to overexpress either p21 or p27 at the time of balloon injury, it was shown that increased expression of the cell cycle inhibitors at an early stage of injury largely prevented neointimal formation. Studies of the role of p21 and p27 in controlling vascular proliferation after systemic injury are ongoing in transgenic mice deficient in these proteins. The relevance of these mechanisms to the pulmonary circulation has not been established. In preliminary studies, it has been found that p27 is constitutively expressed in lung, primarily in airway epithelial cells, but also in smooth muscle cells of large pulmonary arteries. Studies of p21, and of the roles of p21 and p27 in modulating the development of hypoxic pulmonary hypertension, are ongoing. Studies of the embryonal derivation of smooth muscle cells in the development of the arterial tree and the differences in smooth muscle phenotype indicate these cells do not arise from a single type of progenitor cell. Both neuroectoderm and splanchnic mesoderm are sources of smooth muscle cells, the neuroectoderm providing the cells to the vessels of the heart and the thoracic aorta to approximately the middle level of the thoracic aorta. The lower thoracic aorta, abdominal aorta, and leg arteries are supplied by splanchnic mesoderm and the coronary vasculature by cardiac mesenchymal cells. When placed in culture, the neuroectoderm-derived smooth muscle cells grow as polygonal cells in a monolayer, whereas the mesoderm-derived cells grow as spindle-shaped cells that form hills and valleys. All of these cells are responsive to growth factors such as PDGF and show increased thymidine incorporation. However, in response to TGF1, the neuroectoderm-derived cells increased in number, whereas the mesoderm-derived cells did not. In fact, TGF was inhibitory to the mesoderm-derived cells. Only the ectoderm-derived smooth muscle cells expressed the 72-kD gelatinase in response to stimulation with TGF, in contrast to the mesoderm-derived cells. Thus, the differences in embryonic derivation suggest that there also could be lineage-dependent differences, which may help to explain differences in formation of lesions of atherosclerosis and in cell responsiveness in adults, if similar differences exist in humans. These differences were observed in the quail-chick chimeric system and confirmed in the murine system. If they are true for humans, they would have important implications for understanding disease progression in the vessel wall.

In related studies, attempts to understand the smooth muscle cell phenotypes present in the pulmonary artery wall were pursued. Using the bovine pulmonary artery as a model system, four different phenotypes in the media of the bovine pulmonary artery were found. These phenotypes are based on the presence or absence of a number of smooth muscle proteins, including actin, myosin, calponin, desmin, and metavinculin. These differences exist among four identifiable layers present within the media of the artery, which can be separated by dissection and from which cells can be grown in culture. It has been inferred from these different phenotypes that the different proliferative and matrix-producing capabilities of the distinct smooth muscle cell populations observed in culture may govern, at least in part, patterns of abnormal cell proliferation and matrix synthesis in diseases of the vascular system. Within the pulmonary vasculature, it has been observed that isolated medial smooth muscle subpopulations exhibit different proliferative responses to hypoxia, perhaps relevant in the pursuit of mechanisms that regulate responses to low concentrations of oxygen and to understanding the dramatic proliferation of the media because of hyperplastic responses to hypoxia. There were differences in metavinculin in neonatal smooth muscle cells and the adult vessel wall; only the metavinculin smooth muscle cells appear to proliferate in response to injury. By dissecting the different layers from the artery and growing the cells in culture, it has been possible to sort out these different responses and observe differences in cell morphology such as the rounded phenotype cells obtained from the innermost layer of the media and the spindle shape of cells from the outermost layer. Differences in responses to plasma-derived serum and calf serum and differences between signaling molecules in response to growth factor stimulation have also been shown. These differences in signaling molecules represent differences, in part, in G-protein-coupled pathways and responses to cyclooxygenase. Thus, the unique behaviors of different subpopulations of smooth muscle cells may provide useful information regarding the responses of this tissue to injury and the development of arterial disease (13).

The extracellular matrix surrounding smooth muscle cells in the artery wall plays an important role in the control of smooth muscle proliferation. When type I collagen is in fibrillar form, human arterial smooth muscle cells are prevented from traversing the boundary from G2 into S and replicating their DNA. Type I collagen in fibrillar form specifically upregulates the cyclin-dependent kinase inhibitor p27, which binds to a complex of cyclin E, cdk2, and p21. In the absence of p27 binding, this cdk2 molecule is phosphorylated on a threonine molecule, whereas in the presence of p27 this phosphorylation is inhibited, thus preventing cell cycle traverse. These data may be relevant to the observation that smooth muscle cells in the media of the intact artery, which are surrounded by collagen fibrils, are quiescent and nonresponsive to mitogens, whereas smooth muscle cells grown on the same type of collagen in monomeric form grow logarithmically in response to mitogens such as PDGF because of downregulation of p27 and the ability of these cells to traverse the cycle into S phase. This behavior of cultured cells may relate to the fact that fibrillar collagen is absent in the intima of an artery where lesions of atherosclerosis form, and explain why smooth muscle cells in the lesions are responsive in part to mitogens.

The rat, which shows a sexual dimorphism, has been used as a model system to study the effects of estrogen and progestin on smooth muscle proliferation. The well-known inhibition of the development of lesions of atherosclerosis by estrogen in premenopausal women served as the stimulus for these studies. Estrogen affects smooth muscle function (16). The responsivity of vascular smooth muscle cells in culture to estrogen, which inhibits their growth response to mitogens, has been examined. Using various animal models rather than in vitro systems, it has been shown that after endothelial denudation with an intra-arterial balloon catheter, animals with estrogen showed inhibition of smooth muscle cell proliferation as well as inhibition of formation of a neointima in the common and external iliac arteries of the rabbit. Myointimal proliferation after injury to the carotid artery was lower in the intact female Sprague-Dawley rats than in age-matched male rats. It was also demonstrated that this sexual difference is dependent on estrogen rather than on testosterone. Mice, homozygous for deletion of the gene for the classic estrogen receptor, were still protected from a proliferative response by the administration of 17-estradiol, which suggests that the estrogen effect may be mediated by either nongenomic mechanisms or by a novel estrogen receptor. The antiproliferative effects of estrogen in the rat carotid injury model were blocked by the progestin, medroxyprogesterone acetate, without altering 17-estradiol levels, which demonstrates that additional work will be needed before the cellular and molecular mechanisms by which estrogen and progesterone, administered either independently or together, alter smooth muscle cell proliferation.

The protective effects of estrogen against atherosclerosis and the ability of the classic estrogen receptor, when bound to estrogen, to inhibit the development of vascular lesions have been studied in detail. Mice in which this receptor is disrupted and which received carotid artery injury were compared with mice with wild type estrogen receptors. Increases in the neointima and proliferation of smooth muscle cells were quantified and were markedly suppressed in ovariectomized mice treated with physiological levels of 17-estradiol. Neointimal formation in both wild-type and estrogen-receptor-deficient mice suggested that a novel mechanism, independent of the classic estrogen receptor, was involved. These findings led to studies of the recently discovered estrogen receptor  molecule, which is present in human mammary artery and able to activate gene expression in vascular cells. The data are consistent with the presence of the estrogen receptor , which is 80% homologous with the estrogen receptor A (classic estrogen receptor) but differs in its transactivation domain.

	TUMOR-SPECIFIC ANTIGENS AND REGULATORY PROTEINS IN LAM AND TUBEROUS SCLEROSIS COMPLEX

TOP INTRODUCTION CLINICAL ASPECTS OF LAM SMOOTH MUSCLE DIVERSITY,... TUMOR-SPECIFIC ANTIGENS AND... RECOMMENDATIONS REFERENCES

A valuable, comprehensive comparison of tuberous sclerosis complex (TSC) and LAM concluded that there is no evidence that LAM is a form of tuberous sclerosis. The latter was defined as a "congenital hamartomatosis of autosomal dominant inheritance with high penetrance and variable expressivity." It was quickly pointed out, however, that ~ 60% of patients with TSC are apparently sporadic cases with no family history. It was noted that although the signs and symptoms vary widely, pathognomonic lesions of tuberous sclerosis are most often on the skin and in the brain or retina. Lung involvement, reported to occur in fewer than 1% of cases, was present in 2.3% of a series at the Mayo Clinic. Pulmonary lesions, in contrast to the epileptic seizures, usually develop in adult life almost exclusively in women, whereas the incidence of TSC does not differ between the sexes. LAM, on the other hand, is a disease of women, with no known familial association. All characteristics of pulmonary aspects of TSC and LAM are indistinguishable. Understanding their relationship will likely require considerably more specific information concerning derangements of cell function in the two diseases (17).

Evidence was presented that the tuberous sclerosis 2 gene TSC2, one of two genetic loci implicated in TSC, is a tumor-suppressor gene. Tuberin, its protein product, colocalizes in the Golgi with Rap1, which is a Ras-related GTP-binding protein. The deduced 1784-amino acid sequence of tuberin contains a region of similarity to the catalytic domain of Rap1 GTPase-activating protein (GAP). Antibodies against tuberin were used to prepare an immunoprecipitate that displayed Rap1GAP activity. Because tuberin did not increase the GTPase activity of several other Ras-related proteins, the conclusion was that it may function as a specific negative regulator of Rap1. Overexpression of TSC2 in cultured fibroblasts inhibited growth, consistent with a tumor-suppressor function. Expression of TSC2 in cell lines from Eker rats, which are heterozygous for a mutation in the TSC2 gene and predisposed to renal carcinoma, suppressed anchorage-independent growth and tumorigenicity, as expected. It was also found that 50 to 60% of gliomas from patients without TSC lacked detectable tuberin or had elevated amounts of Rap1. All of the findings were consistent with the conclusion that the Rap1 GAP activity of tuberin is physiologically relevant and could explain some of the clinical manifestations of TSC (21). Data suggest that the TSC2 gene may be involved in the pathogenesis of LAM (24).

Both leiomyomas and LAM involve proliferation of smooth-muscle-like cells and are for the most part limited to women of reproductive age. Could LAM and leiomyomas share genetic origins (25)? Estrogen and related hormones increase growth of both lesions. Chromosome abnormalities (13q14 and 14q24) have been noted in both disorders, and a marker for LAM cells (the antigen that reacts with monoclonal antibody HMB45) is located on chromosome 12q13-15, a region frequently rearranged in uterine leiomyomas. To investigate a possible relationship between the two diseases, cells from LAM lungs removed at the same time of transplantation are being grown and clonally expanded in an attempt to develop an in vitro system in which to test the hypothesis that genes "involved in leiomyoma are also involved in LAM."

Many of the melanocyte lineage-specific antigens are enzymes or structural proteins involved in pigment synthesis and melanosome formation that are present in melanoma cells as well as in normal melanocytes (26, 27). It was also noted that malignant melanoma is one of the most immunogenic human tumors known. One of the antigens, gp100, of unknown function, is variably glycosylated. A gp100 cDNA clone provided deduced sequence of a 661-amino acid transmembrane protein; it has a short C-terminal domain in the cytoplasm and the largest part of the protein inside the melanosome. Immunoreactive gp100 was detected as premelanocyte structures before their fusion with enzyme-containing packages. Of several anti-gp100 monoclonal antibodies, two react with melanoma cells but not with melanocytes. One of these, HMB45, reacts also with melanosomelike structures in LAM cells. A protein from human melanocytes termed Pmel 17 is 98% identical to gp100. The two proteins result from alternative splicing of transcripts from the same gene. Because tumor-infiltrating lymphocytes from patients with melanoma react with peptides derived from these proteins, the peptides are being used with cells from healthy donors to generate specific cytotoxic lymphocytes in an attempt to learn whether gp100 could be a useful target for specific immunotherapy of malignant melanoma.

Results of cellular and molecular studies of 37 patients with LAM seen at the NIH Clinical Center during the past 18 mo were summarized (28). The gp100 mRNA was detected in lung from patients with LAM, but not in normal lung. On electron microscopy, gp100 immunoreactivity was confined to small electron-dense granules with a fine lamellar structure that resemble immature melanosomes. Double staining with HMB45 and an antibody against proliferating cell nuclear antigen (PCNA) identified LAM cells that reacted with one or the other or both and differed in morphology. PCNA was present in 25 to 76% of LAM cells and < 5% of normal bronchial or vascular smooth muscle cells. After evaluation of expression of genes of the Bcl-2 family that regulate apoptosis, it was suggested that LAM cells have high mitotic and low apoptotic activities that are probably important in their proliferative behavior. There was an inverse relationship between cell proliferation (PCNA content) and amount of gp100. CA125, a marker protein for nonmucinous epithelial ovarian cancer, is found in the serum of some LAM patients. Levels of CA125 were elevated in only six of 37 patients. All had ascites or chylothorax. CA125 was detected by immunofluorescence in the pleura of patients with elevated serum CA125, but not those with normal levels. No CA125 was detected in LAM cells whether or not the patient had elevated CA125 in the serum. The conclusion was that circulating CA125 in patients with LAM with chylothorax or ascites is most likely derived from mesothelial cells of the pleura and peritoneum. It was commented that serum CA125 is also elevated in patients with pleural effusion without LAM.

Several topics discussed earlier were reconsidered. Among the more obviously intriguing, and potentially relevant to LAM, were studies of smooth muscle cell heterogeneity and diversity of lineage. The description of development of mammalian vascular smooth muscle cells from at least four different embryonic sources could be useful in elucidating the genesis of LAM cells. Additionally, demonstration that arteries can be composed of smooth muscle cells derived from different progenitors that are functionally diverse in their responses to growth factors, hypoxia, or injury, as well as in the matrix protein synthesis and proliferative capacity, suggests new approaches to elucidating the life history of LAM cells. Finally, the reports of effects of estrogen on smooth muscle proliferation were clearly timely. The data describing a role of the relatively recently recognized estrogen receptor  in estrogen effects on vascular smooth muscle emphasize the potential importance of characterizing estrogen receptors and effects in LAM cells.

	RECOMMENDATIONS

TOP INTRODUCTION CLINICAL ASPECTS OF LAM SMOOTH MUSCLE DIVERSITY,... TUMOR-SPECIFIC ANTIGENS AND... RECOMMENDATIONS REFERENCES

1. Establish diagnostic criteria for LAM. The diagnostic criteria for LAM may not necessarily include a lung biopsy, especially in view of the specificity and sensitivity of a high resolution CT scan. The criteria for inclusion of patients in the LAM registry may differ, however, from those used for diagnosis of patients by individual pulmonologists.
2. Encourage exchange of registry information among countries. In view of the studies being performed on this disease in Europe, the Far East, and North America, it would be useful to have information exchange among countries.
3. Encourage collaborations among several laboratories to facilitate elucidation of the lineage of the smooth muscle cells in LAM and the origin of the melanoma proteins.
4. Characterize specific estrogen receptors in LAM cells. Newer estrogen analogues might be helpful adjuncts to therapy of the disease, and possibly lead to fewer side effects. What is the role of specific estrogen receptors in LAM cells? Can the selection of a progesterone analogue for treatment be optimized further to minimize side effects?
5. Define the association of LAM with tuberous sclerosis complex (TSC) in more detail as more information is obtained regarding the TSC genes and their effects on cell proliferation.
6. Investigate further the basis for the recurrence of LAM in patients after lung transplantation. Is it secondary to a circulating factor (e.g., growth factor)? What is the rate of recurrence? Should patients after transplant be treated with progesterone?

	Footnotes

Correspondence and requests for reprints should be addressed to Carol E. Vreim, Ph.D., Division of Lung Diseases, National Heart, Lung, and Blood Institute, Rockledge 2, Suite 10018, 6701 Rockledge Dr., MSC 7952, Bethesda, MD 20892-7952.

(Received in original form March 23, 1998 and in revised form October 8, 1998).

	References

TOP INTRODUCTION CLINICAL ASPECTS OF LAM SMOOTH MUSCLE DIVERSITY,... TUMOR-SPECIFIC ANTIGENS AND... RECOMMENDATIONS REFERENCES

1. Kitaichi, M., K. Nishimura, H. Itoh, and T. Izumi. 1995. Pulmonary lymphangioleiomyomatosis: a report of 46 patients including a clinicopathologic study of prognostic factors. Am. J. Respir. Crit. Care Med. 151: 527-533 [Abstract].

2. Cordier, J.-F., R. Lazor, and Groupe d'Etudes et de Recherche sur les Maladies "Orphelines" Pulmonaires. 1998. Perspectives on lymphangioleiomyomatosis in France. In J. Moss, editor. LAM and Other Diseases Characterized by Smooth Muscle Proliferation. Marcel Dekker, New York, NY. 9-31.

3. Lipman, M. C. I., and R. M. DuBois. 1998. Clinical experience with lymphangioleiomyomatosis in the United Kingdom. In J. Moss, editor. LAM and Other Diseases Characterized by Smooth Muscle Proliferation. Marcel Dekker, New York, NY. 33-43.

4. Kalassian, K. G., R. Doyle, P. Kao, S. Ruoss, and T. A. Raffin. 1997. Lymphangioleiomyomatosis: new insights. Am. J. Respir. Crit. Care Med. 155: 1183-1186 [Medline].

5. Crausman, R. S., F. X. McCormack, and T. E. King, Jr. 1998. Lymphangioleiomyomatosis: differential diagnosis. In J. Moss, editor. LAM and Other Diseases Characterized by Smooth Muscle Proliferation. Marcel Dekker, New York, NY. 111-141.

6. Avila, N. A., C. C. Chen, S. C. Chu, and J. Moss. 1998. Lymphangioleiomyomatosis: imaging features. In J. Moss, editor. LAM and Other Diseases Characterized by Smooth Muscle Proliferation. Marcel Dekker, New York, NY. 155-169.

7. Travis, W. D., J. Usuki, K. Horiba, and V. Ferrans. 1998. Histopathologic studies on lymphangioleiomyomatosis. In J. Moss, editor. LAM and Other Diseases Characterized by Smooth Muscle Proliferation. Marcel Dekker, New York, NY. 171-217.

8. Fukuda, Y. 1998. Ultrastructural pathology of pulmonary lymphangioleiomyomatosis. In J. Moss, editor. LAM and Other Diseases Characterized by Smooth Muscle Proliferation. Marcel Dekker, New York, NY. 219-239.

9. Trulock, E. P. 1988. Lung transplantation: special considerations and outcome in LAM. In J. Moss, editor. LAM and Other Diseases Characterized by Smooth Muscle Proliferation. Marcel Dekker, New York, NY. 65-78.

10. Boehler, A., R. Speich, E. W. Russi, and W. Weder. 1998. International survey on lung transplantation in Lymphangioleiomyomatosis. In J. Moss, editor. LAM and Other Diseases Characterized by Smooth Muscle Proliferation. Marcel Dekker, New York, NY. 79-98.

11. Sullivan, E. J., and J. K. Stoller. 1998. The registry for individuals with Lymphangioleiomyomatosis. In J. Moss, editor. LAM and Other Diseases Characterized by Smooth Muscle Proliferation. Marcel Dekker, New York, NY. 99-110.

12. Duckers, H. J., B. Manfred, E. G. Nabel, and Z. Y. Yang. 1998. Regulation of cell cycle progression in vascular smooth muscle cells: breaking the cycle. In J. Moss, editor. LAM and Other Diseases Characterized by Smooth Muscle Proliferation. Marcel Dekker, New York, NY. 241-253.

13. Frid, M. G., A. A. Aldashev, E. C. Dempsey, and K. R. Stenmark. 1997. Smooth muscle cells isolated from discrete compartments of the mature vascular media exhibit unique phenotypes and distinct growth capabilities. Circ. Res. 81: 940-952 [Abstract/Free Full Text].

14. Chamley-Campbell, J., G. Campbell, and R. Ross. 1981. Phenotype- dependent response of cultured aortic smooth muscle to serum mitogens. J. Cell. Biol. 89: 379-383 [Abstract/Free Full Text].

15. Topouzis, S., and M. W. Majesky. 1996. Smooth muscle lineage diversity in the chick embryo: two types of aortic smooth muscle cell differ in growth and receptor-mediated transcriptional responses to TGF-1. Dev. Biol. 178: 430-445 .

16. Speir, E., and R. O. Cannon, III. 1998. Role of estrogen in smooth muscle growth and function. In J. Moss, editor. LAM and Other Diseases Characterized by Smooth Muscle Proliferation. Marcel Dekker, New York, NY. 351-371.

17. Castro, M., C. W. Shepherd, M. R. Gomez, J. T. Lie, and J. H. Ryu. 1995. Pulmonary tuberous sclerosis. Chest 107: 189-195 [Abstract/Free Full Text].

18. Kerr, L. A., M. L. Blute, J. H. Ryu, S. J. Swensen, and R. S. Malek. 1993. Renal angiomyolipoma in association with pulmonary lymphangioleiomyomatosis: forme fruste of tuberous sclerosis? Urology 41: 440-444 [Medline].

19. Short, M. P., and D. J. Kwiatkowski. 1998. Clinical and genetic aspects of tuberous sclerosis complex I and II. In J. Moss, editor. LAM and Other Diseases Characterized by Smooth Muscle Proliferation. Marcel Dekker, New York, NY. 387-406.

20. Ryu, J. H. 1998. Tuberous sclerosis complex and LAM. In J. Moss, editor. LAM and Other Diseases Characterized by Smooth Muscle Proliferation. Marcel Dekker, New York, NY. 407-418.

21. Wienecke, R., A. Konig, and J. E. DeClue. 1995. Identification of tuberin, the tuberous sclerosis-2 product: tuberin possesses specific Rap1GAP activity. J. Biol. Chem. 270: 16409-16414 [Abstract/Free Full Text].

22. Wienecke, R., J. C. Maize, J. Reed, J. deGunzburg, R. S. Yeung, and J. E. DeClue. 1997. Expression of the TSC2 product tuberin and its target Rap1 in normal human tissues. Am. J. Pathol. 150: 48-50 .

23. DeClue, J. E. 1998. Tuberin. In J. Moss, editor. LAM and Other Diseases Characterized by Smooth Muscle Proliferation. Marcel Dekker, New York, NY. 419-440.

24. Smolarek, T. A., L. L. Wessner, F. X. McCormack, J. C. Mylet, A. G. Menon, and E. P. Henske. 1998. Evidence that lymphangiomyomatosis is caused by TSC2 mutations: chromosome 16p13 loss of heterozygosity in angiomyolipomas and lymph nodes from women with lymphangiomyomatosis. Am. J. Hum. Genet. 62: 810-815 [Medline].

25. McCormack, F. X., T. A. Smolarek, and A. G. Menon. 1998. Lymphangioleiomyomatosis and leiomyoma: possibility of shared genetic origins? In J. Moss, editor. LAM and Other Diseases Characterized by Smooth Muscle Proliferation. Marcel Dekker, New York, NY. 373- 386.

26. Spritz, R. A., and V. J. Hearing Jr.. 1994. Genetic disorders of pigmentation. Adv. Hum. Genet. 22: 1-45 [Medline].

27. Schreurs, M. W. J., and C. G. Figdor. 1998. Molecular and immunological characterization of melanocyte lineage-specific antigens. In J. Moss, editor. LAM and Other Diseases Characterized by Smooth Muscle Proliferation. Marcel Dekker, New York, NY. 479-499.

28. Chu, S. C., K. Horiba, J. Usuki, N. A. Avila, C. C. Chen, W. D. Travis, V. J. Ferrans, and J. Moss. 1997. Comprehensive evaluation of 25 patients with lymphangioleiomyomatosis (LAM) (abstract). Am. J. Respir. Crit. Care Med. 155: A330 .

29. Usuki, J., K. Horiba, Y. Matsumoto, S. C. Chu, V. J. Ferrans, and J. Moss. 1998. Inverse correlation between markers of cell proliferation and expression of melanosomal antigens in smooth muscle cells in lymphangioleiomyomatosis (LAM) and in cultured melanoma cell lines (abstract). Am. J. Respir. Crit. Care Med. 157: A806 .

30. Kelly, J. A., S. C. Chan, and J. Moss. 1998. Cellular heterogeneity in lymphangioleiomyomatosis. In J. Moss, editor. LAM and Other Diseases Characterized by Smooth Muscle Proliferation. Marcel Dekker, New York, NY. 501-516.