Is the distribution of dermal neurofibromas in neurofibromatosis type 1 (NF1) related to the pattern of the skin surface temperature?

ARTICLE

The formation of numerous dermal neurofibromas is a hallmark of neurofibromatosis type 1 (NF1) [1-3]. These tumours differ in biology from the superficially visible diffuse plexiform neurofibromas: Plexiform neurofibromas are thought to be congenital lesions occurring relatively infrequently [1, 4]. NF1 has been shown to have tumour suppressor gene function. According to Knudson's "two hit model" a total loss of the NF1 product neurofibromin in precursor cells should precede the tumourigenesis. Haplo insufficiency is the consequence of most constitutional NF1 mutations [5]. Subsequent mutations in the second NF1 allele (wild type allele) have been detected in dermal neurofibromas [6-10]. The occurrence of dermal neurofibromas is thought to be caused exclusively by stochastic inactivation of the wild type allele in somatic cells. In this case, one can predict that first the number of dermal neurofibromas increases progressively with age, and second that the dermal neurofibromas are randomly distributed according to the distribution of their precursor cells. Here, we investigated whether these predictions are able to describe the distribution of dermal neurofibromas in our NF1 patients. We found obvious differences between these predictions and the observed pattern. We suggest that additional factors, especially the skin surface temperature, influence the occurrence of dermal neurofibromas.

Methods

In total, more than 600 NF1 patients were carefully examined clinically according to the NIH criteria [11] using a detailed protocol. This was done in the Departments of Medizische Genetik (Berlin) and Humangenetik, the NF-Ambulanz (Ulm) and the Department of Laser-Medizin (Berlin). The specific NF1 mutation was identified in several of these patients [5]. Thermographs were done in the Department of Lasermedizin using a nitrogen cooled thermograph camera (Varioscan, Jenoptik, Germany) under standardized conditions.

Results

Does the number of dermal neurofibromas increase with age?

Generally, it is accepted that the number of dermal neurofibromas increases progressively after their first clinical appearance, usually during puberty [1, 2]. This is confirmed by our clinical investigations on more than 600 patients (0.1 to 82 in age) for over more than 10 years. The increase in number per year differs between the patients, but we do not know of any adult NF1 patient with a stable number of neurofibromas over a period of 5 years. This progressive increase fits the first prediction in which stochastic somatic NF1 mutations cause these tumours. But two clinical observations by others and us do not fit this model. Firstly, the increase in number of dermal neurofibromas up to the age of 30-45 is followed by a "relatively quiescent" period [1]. Secondly, pregnancy leads to an increase in the number and size of dermal neurofibromas [1, 12]. All female NF1 patients we examined fitted in with this observation. An additional indication in this context is that the diagnosis of sporadic NF1 is often made for the first time during pregnancy. Thus, additional mechanisms seem to influence the occurrence of the dermal neurofibromas.

Is there a stochastic distribution of dermal neurofibromas in NF1?

Dermal neurofibromas may occur everywhere on the body surface. However, they are found preferentially on the trunk, their number decreasing towards the body periphery as described [2, 13]. This is confirmed by our clinical investigation of more than 600 patients. An example is shown in Figure 1. Apart from this distribution, some skin areas are relatively free of dermal neurofibromas, e.g. the nose and the ears. This is observed in all severely affected NF1 patients we clinically examined with a large number of facial neurofibromas. A typical example is shown in Figure 2. Other locations are the plantar surface of the foot and the hairy skull. These observations suggest that dermal neurofibromas are not randomly distributed over the body surface as predicted in the model.

Pattern of dermal neurofibromas and density of sensitive nerves in the dermis

One must ask how these preferential localisations can be explained. The difference in number between the trunk and the periphery resembles the differences in density of free intraepidermal nerve endings described recently [14, 15]. There, it was demonstrated that the number of these nerve endings decreases from the trunk to the distal parts of the limbs. Dermal neurofibromas should originate from terminal nerve branches in the skin [16]. The Schwann cells are favoured as progenitor cells [8, 10] and not the perineural, endoneural or epineural cells also found aside from mast cells in these tumours [16-18]. Until now, the density distribution of Schwann cells of the sensitive nerves is not known in detail. Therefore, it is not possible to examine whether or not a numerical correlation exits between the pattern of Schwann cells of sensitive nerves and dermal neurofibromas.

Pattern of dermal neurofibromas and body surface temperature

Prima vista, the normal pattern of the body surface temperature [19] correlates to the occurrence of dermal neurofibromas. The highest temperature (35° C) is found on the trunk whereas the arms and the legs are significantly cooler (28° C). The face, the nose and the ears are the coldest regions. The NF1 patients examined until now by a thermograph, share principally the same distribution of body surface temperature as normal human beings (see as example Fig. 3). If one compares this pattern with the distribution of the dermal neurofibromas on the trunk and periphery, it is obvious that the tumours occur predominantly in regions with a higher surface temperature (Fig. 4). In addition, the nose and the ears in NF1 patients are clearly colder than the face (Fig. 5). In these regions the number of neurofibromas is reduced, as shown in Figure 2. During pregnancy, the average body temperature increases by about 1° C. In addition, dermal neurofibromas increase in number and size during this time in all skin areas. Taken together, we assume that the average temperature of the body surface is involved in the development of dermal neurofibromas.

Discussion

In NF1, the pattern of dermal neurofibromas is not very well understood. It is not related to the classical pattern described for mosaics [21]. It is proposed that the occurrence of these tumours is directly correlated to the genetic loss of the second NF1 allele. If so, the dermal pattern of these tumours should represent the local somatic NF1 mutation rate. The NF1 germline mutation rate is very high [3]. The somatic NF1 mutation rate has not yet been investigated. It remains to be examined whether the somatic NF1 mutation rate is increased by exogenous factors such as mutagenes. The nose is drastically exposed to mutagenic UV radiation but is relatively free of neurofibromas, in contrast to the tumour pattern observed in xeroderma pigmentosum. There are three possible explanations that might be made to explain this observation. Firstly, that UV radiation does not influence the NF1 mutation rate, which seems unlikely. Secondly, that no precursor cells for neurofibromas exist in the dermis of the nose. But this is not true. Thirdly, that the second NF1 mutation alone is not a sufficient condition for the formation of neurofibromas. We prefer the third explanation. Additional factors seem to be important for the local formation of a dermal neurofibroma. Here, we hypothesise that the surface temperature should be considered as a modifier for the pattern of dermal neurofibromas. Whether the surface skin temperature is related directly to the density of sensitive nerves remains unclear. It has been shown in temperature-dependent sex determination in reptiles that a small difference in the external temperature (31.5 versus 32.5° C) is able to determine the phenotype completely [22, 23]. We do not know if this ancient evolutionary way of regulating a differentiation is conserved in regulations in human skin. How could the body surface temperature influence the occurrence of dermal neurofibromas? We speculate, firstly by reducing the amount of neurofibromin. In areas of higher body surface temperature mechanisms functionally inactivating the NF1 wild type allele may be more active than in cooler areas, such as seen in cases of somatic mutagenesis. It was shown for the HPRT gene in cultured fibroblasts that a higher temperature increases the mutation rate. In the presence of azaserine a rise in temperature from 33 to 37 degrees leads to a more than 10-fold increase in mutation rate per cell generation [24]. A clinical example for the temperature dependence of mutation rate may be the increased predisposition to tumours of the testis in cryptorchidism. Also, the stability of the NF1 mRNA [25], of neurofibromin [26, 27] or of the NF1 RNA editing [28] may be temperature sensitive. Recently, an in vivo temperature-sensitive defect of transcription and DNA repair due to thermo-instability of TFIIH, a DNA repair/transcription factor, was described in patients with trichothiodystrophy [29]. Secondly, by influencing mechanisms involved indirectly in neurofibroma development. It is to be noticed that the majority of the neurofibroma cells still have the NF1 wild type allele and the second NF1 hit is detectable only in a minority of cells, and only in a minority of the Schwann cells as demonstrated recently [10]. It is not yet understood how the NF1 haploinsufficient (NF1^+/-) cells modify the behaviour of the NF1 null (NF1^-/-) cells regarding the induction of the dermal neurofibromas, and if any of these mechanisms is temperature sensitive [20]. How the local skin temperature is correlated to the pattern of dermal neurofibromas remains to be examined in more detail on the clinical level, for instance in patients with altered temperature regulation like patients with hyper- or hypothyroidism. Above all, it will be interesting to investigate whether the intrafamilial variability in the number of dermal neurofibromas, suggested to be related to modifying genes [30, 31], correlates with differences in the main skin temperature.

CONCLUSION

We thank B. Bartelt, G. Vergani, H.-P. Berlien and W. Vogel for helpful discussions, B. Jamil for critically reading the manuscript and A. Großewinkelmann and L. Weinberg for performing the thermographs.

REFERENCES

1. Riccardi VM. Neurofibromatosis: Phenotype, Natural History, and Pathogenesis, Baltimore: 2nd Edition, John Hopkins University Press, 1992.

2. Huson SM. Neurofibromatosis 1: a clinical and genetic overview. In: Huson SM and Hughes RAC, ed. The neurofibromatoses: a pathogenetic clinical overview. London: Chapman & Hall, 1994: 160-3.

3. Upadhyaya M, Cooper DN. Neurofibromatosis type 1 from genotype to phenotype. Oxford: BIOS Scientific Publisher Ltd, 1998.

4. Korf BC. Plexiform neurofibromas. Am J Med Genet 1999; 89: 31-7.

5. Fahsold R, Hoffmeyer S, Mischung C, Gille C, Ehlers C, Kucukceylan N, Abdel-Nour M, Gewies A, Peters H, Kaufmann D, Buske A, Tinschert S, Nuernberg P. Minor lesion mutational spectrum of the entire NF1 gene does not explain its high mutability but points to a functional domain upstream of the GAP-related domain. Am J Hum Genet 2000; 66: 790-18.

6. Colman SD, Williams CA, Wallace MR. Benign neurofibromas in type 1 neurofibromatosis (NF1) show somatic deletions of the NF1 gene. Nat Genet 1995; 11: 90-2.

7. Sawada S, Florell S, Purandare SM, Ota M, Stephens K, Viskochil DH. Identification of NF1 mutations in both alleles of a dermal neurofibroma. Nat Genet 1996; 14: 110-2.

8. Serra E, Puig S, Otero D, Gaona A, Kruyer H, Ars E, Estivill X, Lazaro C. Confirmation of a double-hit model for the NF1 gene in benign neurofibromas. Am J Hum Genet 1997; 61: 512-9.

9. Kluwe L, Friedrich R, Mautner VF. Loss of NF1 allele in Schwann cells but not in fibroblasts derived from an NF1-associated neurofibroma. Genes Chromosomes Cancer 1999; 24: 283-5.

10. Serra E, Rosenbaum T, Winner U, Aledo R, Ars E, Estivill X, Lenard HG, Lazaro C. Schwann cells harbor the somatic NF1 mutation in neurofibromas: evidence of two different Schwann cell subpopulations. Hum Mol Genet 2000; 9: 3055-64.

11. Gutmann DH, Aylsworth A, Carey JC, Korf B, Marks J, Pyeritz RE, Rubenstein A, Viskochil D. The diagnostic evaluation and multidisciplinary management of neurofibromatosis 1 and neurofibromatosis 2. JAMA 1997; 278: 51-7.

12. Dugoff L, Sujansky E. Neurofibromatosis type 1 and pregnancy. Am J Med Genet 1996; 66: 7-10.

13. Siemens HW. Klinisch-dermatologische Studien über die Recklinghausensche Krankheit. Arch Derm Syph 1926; 150: 80-3.

14. Lauria G, Holland N, Hauer P, Cornblath DR, Griffin JW, McArthur JC. Epidermal innervation: changes with aging, topographic location, and in sensory neuropathy. J Neurol Sci 1999; 164: 172-8.

15. Johansson O, Wang L, Hilliges M, Liang Y. Intraepidermal nerves in human skin: PGP 9.5 immunohistochemistry with special reference to the nerve density in skin from different body regions. J Peripher Nerv Syst 1999; 4: 43-52.

16. Wiestler OD, Radner H. Pathology of neurofibromatosis 1 and 2. In: Huson SM, Hughes RAC, ed. The Neurofibromatoses, London: Chapman & Hall, 1994: 135-59.

17. Peltonen J, Aho H, Halme T, Nanto-Salonen K, Lehto M, Foidart J M, Duance V, Vaheri A, Penttinen R. Distribution of different collagen types and fibronectin in neurofibromatosis tumours. Acta Pathol Microbiol Immunol Scand [A] 1984; 92: 345-52.

18. Peltonen J, Jaakkola S, Lebwohl M, Renvall S, Risteli L, Virtanen I, Uitto J. Cellular differentiation and expression of matrix genes in type 1 neurofibromatosis. Lab Invest 1988; 59: 760-71.

19. Guyton AC, Hall JE. Textbook of Medical Physiology. CO: 10th ed, WB Saunders, 2000.

20. Rosenbaum T, Patrie KM, Ratner N. Neurofibromatosis type 1: genetic and cellular mechanisms of peripheral nerve tumour formation. Neuroscientist 1997; 3: 412-20.

21. Happle R. Mosaicism in human skin. Understanding the patterns and mechanisms. Arch Dermatol 1993; 129: 1460-70.

22. Lang JW, Andrews HV. Temperature-dependent sex determination in crocodilians. J Exp Zool 1994; 270: 28-44.

23. Western PS, Harry JL, Graves JA, Sinclair AH. Temperature-dependent sex determination in the american alligator: AMH precedes SOX9 expression. Dev Dyn 1999; 216: 411-9.

24. Simons JW. Effect of temperature on mutation in cultured human skin fibroblasts. Mutat Res 1982; 22: 417-26.

25. Cowley GS, Murthy AE, Parry DM, Schneider G, Korf B, Upadhyaya M, Harper P, MacCollin M, Bernards A, Gusella JF. Genetic variation in the 3' untranslated region of the neurofibromatosis 1 gene: application to unequal allelic expression. Somat Cell Mol Genet 1998; 24: 107-19.

26. Griesser J, Kaufmann D, Maier B, Mailhammer R, Kuehl P, Krone W. Post-transcriptional regulation of neurofibromin level in cultured human melanocytes in response to growth factors. J Invest Dermatol 1997; 108: 275-80.

27. Kaufmann D, Bartelt B, Hoffmeyer S, Müller R. In cultured melanocytes derived from café au lait macules of neurofibromatosis type 1 patients neurofibromin content can be regulated by alteration of the neurofibromin half-life. Arch Dermatol Res 1999; 291: 312-7.

28. Cappione AJ, French BL, Skuse GR. A potential role for NF1 mRNA editing in the pathogenesis of NF1 tumours. Am J Hum Genet 1997; 60: 305-12.

29. Vermeulen W, Rademakers S, Jaspers NG, Appeldoorn E, Raams A, Klein B, Kleijer WJ, Hansen LK, Hoeijmakers JH. A temperature-sensitive disorder in basal transcription and DNA repair in humans. Nat Genet 2001; 27: 299-303.

30. Easton DF, Ponder MA, Huson SM, Ponder BA. An analysis of variation in expression of neurofibromatosis (NF) type 1 (NF1): evidence for modifying genes. Am J Hum Genet 1993; 53: 305-13.

31. Carey JC, Viskochil DH. Neurofibromatosis type 1: a model condition for the study of the molecular basis of variable expressivity in human disorders. Am J Med Genet 1999; 89: 7-13.