THE SOTOS SYNDROME - NSD1 HAPLOINSUFFICIENCY: CEREBRAL GIGANTISM UPDATE

« Back to Volume 22, Issue 3, September 2006 - Table of Contents

Allen W. Root, MD 1,2
Frank B. Diamond, Jr., MD 1
1 Departments of Pediatrics and 2Biochemistry and Molecular Biology
University of South Florida College of Medicine, Tampa, FL
All Children’s Hospital, St. Petersburg, FL

INTRODUCTION

Cerebral gigantism (OMIM 117550) is characterized by excessive pre- and postnatal growth, a characteristic face, and developmental delay with a prevalence of ~1:14,000 births.1 In more than 90% of patients, Sotos syndrome is due to haploinsufficiency of NSD1 (Nuclear Receptor-Binding Su-var, Enhancer of Zeste, and Trithorax Domain Protein 1, chromosome 5q35, OMIM 606681).2 Cerebral gigantism is thus a genomic disorder–a pathologic state due to loss, gain, or disruption of a dosage-sensitive gene that results in a recognized phenotype.3

CLINICAL CHARACTERISTICS

In patients with cerebral gigantism, rapid linear growth begins during gestation; at birth, length is more than +2 standard deviations (SD) above mean length for gestational age and gender in 85% of neonates, while birth weights are usually within the high normal range. Neonates with Sotos syndrome may also have prolonged jaundice, hypotonia, and feeding difficulties.2 Linear growth remains rapid throughout infancy and childhood. Because skeletal maturation is also advanced, adult heights of patients with Sotos syndrome are usually near or slightly above +2 SD; however, adult stature usually exceeds target height by an average of 11 cm in males and 6 cm in females (Figure 1).

Figure 1.
A 22 year old male with cerebral gigantism.

A 22 year old male with cerebral gigantism.

Occipito-frontal head circumference is also increased during infancy and childhood and remains above the 97th percentile in most adults with Sotos syndrome. However, in 10% of subjects with cerebral gigantism, height and head circumference remain within the normal ranges.4 The “acromegalic-like” face of the patient with cerebral gigantism is characterized by a high, broad, and bossed forehead with sparse fronto-temporal hair, long and narrow face, down-slanting of the palpebral fissures, malar flushing, and a sharply pointed, prognathic mandible that becomes more evident over time.5,6 The palate is highly arched; hands and feet are prominent; and scoliosis occurs frequently. Subjects with Sotos syndrome also have anomalies of the cardiovascular (patent ductus arteriosus, atrial septal defect), genitourinary (agenesis, duplication, vesicoureteral reflux, hypospadias, cryptorchidism), and central nervous systems (hypoplasia of the corpus callosum, ventricular dilatation, enlarged extracerebral fluid spaces); electroencephalographic abnormalities and seizures occur in some subjects. Most but not all patients with cerebral gigantism have mild to severe developmental delay, a problem that may ameliorate somewhat as the patient ages.7 In addition, Sotos syndrome patients may develop aggressive behavior and may manifest psychoses in adulthood.8,9 Neoplasms develop in 2% to 4% of children with cerebral gigantism, including acute lymphoblastic leukemia, T-cell lymphoma, Wilms tumor, sacrococcygeal teratoma, presacral ganglioneuroma, hepatocellular carcinoma, neuroblastoma, and ganglioglioma.10,11

PATHOGENESIS

The mechanism(s) underlying the extreme growth of patients with Sotos syndrome is unknown. Growth hormone secretion is normal in patients with cerebral gigantism; serum concentrations of insulin-like growth factor (IGF)-I and acid labile subunit have been normal, a bit low, or somewhat elevated in various reports.12 In some subjects, serum levels of IGF-II and IGF-binding proteins (IGFBP)-3 and -4 have been within the low normal range. The rate of IGFBP-3 proteolysis was accelerated in one report, suggesting that free IGF-I values might be increased in this disorder.12 Prostate specific antigen (PSA) is one of several proteolytic enzymes that degrade IGFBP-3.13,14 Perhaps measurements of PSA levels in patients with cerebral gigantism would be of interest and contribute to our understanding of the pathogenesis of this disorder.

GENETICS

Heterozygous microdeletions and loss-of-function mutations in NSD1 resulting in haploinsufficiency of the gene product have been identified in more than 90% of patients with Sotos syndrome.2,4,5NSD1 contains 23 exons and encodes a 2596 to 2696 amino acid, broadly expressed protein (brain, muscle, kidney, spleen, thymus, lymph node, lung) that functions as a co-regulator of transcription by interacting with nuclear transcription factors and as a histone methyltransferase.11 Within the structure of NSD1, there is a SET domain, a conserved sequence of approximately 150 amino acids that remodels chromatin structure by histone methylation, thereby modulating gene transcription; NSD1 specifically methylates lysine-36 of histone H3 and lysine-20 of histone-H4.15 By inserting its side chain into a cleft within the SET-containing protein, the selected histone H3/4-lysine accesses both the enzymatic site and its methyl donor S-adenosyl-L-methionine.16 Figure 2 shows a structure of the SET 7/9 ternary complex. This structure is highly specific for the histone methyltransferase target. Methylation of one or the other histone H3/4-lysine residues usually, but not necessarily always, exerts an inhibitory (silencing) effect on the transcription of a targeted gene. Encoded within NSD1 are several additional functional domains, including one SET-associated cysteine-rich domain, two nuclear receptor interactive domains (exon 2), two proline-tryptophan-tryptophan-proline (PWWP) domains (exons 3-4, 15-17), and 5 plant homeodomains (PHD) (exons 11-17, 22).5,6 A PHD has a zinc-finger structure that permits interaction with chromatin, while the PWWP domains are involved in protein-protein interaction.

Figure 2.

Figure 2.

Structure of the SET 7/9 ternar y complex. a, Two or thogonal views of the SET 7/9 ternary complex in ribbons representation. The N-terminal domain is colored pink, the SET domain is blue and the C-terminal segment is grey. The H3 peptide is indicated in green, with the side chain of methylated Lys 4 shown. The S-adenosyl-L-homocysteine (AdoHcy) cofactor is colored yellow. The secondar y structure elements are labeled according to our earlier structure. Two small turns of the 310 helix are also labeled. b, Two views of the SET domain are shown in a surface representation colored according to electrostatic potential (the two views are related by a twofold rotation about a vertical axis). The left panel shows AdoHcy colored yellow; the right panel shows the H3 peptide colored green. The inset panel shows a close-up view of the lysine access channel containing the methyl lysine side chain as viewed from the S-adenosyl-L-methionine(AdoMet)-binding site.

Reprinted with permission: Xiao B, et al. Nature. 2003; 421:652–656. Copyright © Nature Publishing Group. 2003. All rights reserved.

NSD1 binds to transcription factors and co-factors where it may behave as either a co-activator or co-repressor, depending on which of its two nuclear interactive domains is involved.5,11 By binding to the intact or carboxyl terminal region of the nuclear androgen receptor (AR) through its activating nuclear receptor interactive domain, NSD1 acts as a co-regulatory factor that enhances AR transcriptional activity. NSD1 also interacts with NIZP1, a zinc-finger DNA-binding repressor of transcription that directs the histone methyltransferase SET domain of NSD1 to targeted gene promoters. 17NSD1 and other SET-domain containing proteins interact with factors that regulate cell growth and have been implicated in several human malignancies such as acute myeloid leukemia of childhood. 18 However, the mechanism(s) by which inactivating mutations of NSD1 lead to the clinical manifestations of Sotos syndrome remains unclear at present. Since homozygous NSD1 -/- mutant mouse embryos succumb very early in gestation, NSD1 is also crucial for early post-implantation mammalian fetal development.13 However, the NSD1 +/- mutant mouse is phenotypically normal.

Microdeletions of 1.9Mb-encompassing NSD1 are the most common mutations identified in subjects with cerebral gigantism of Japanese ancestry19 (Figure 3). Mechanistically important in the process that leads to microdeletions of NSD1 in this population is non-allelic homologous recombination or unequal rearrangement of low-copy repeat sequences that flank the distal and proximal breakpoints that encompass NSD1.4,20,21 Preferentially, microdeletions of NSD1 are of paternal origin in Japanese patients with Sotos syndrome. Their fathers have a heterozygous inversion of nucleotides flanking NSD1 on chromosome 5 that predisposes to unequal intrachromosomal recombination during meiosis that leads to microdeletions (or duplications) in their progeny.11,21,22 In western populations, microdeletions of NSD1 are variable in size, due to interchromosomal rearrangement, and far less frequent, accounting for less than 10% of the identified NSD1 mutations in patients with Sotos syndrome. More than 100 intragenic inactivating splice site, frameshift (due to small insertions and deletions), nonsense, and missense NSD1 mutations that account for more than 90% of the genetic abnormalities identified in non-Japanese patients with cerebral gigantism have been identified.5,23 Missense mutations are clustered between exons 13 and 23 within conserved functional domains.24

Figure 3.
Figure 3.

NSD1 mutations in individuals with Sotos syndrome. a, BAC/PAC/cosmid map spanning the 5q35 breakpoint. Red and blue horizontal lines indicate clones spanning the breakpoint (detected by FISH analysis) and complete genomic sequences, respectively. Arrows indicate genes,and green boxes below NSD1 represent exons 1,2 and 3. b, Genomic structure of NSD1 and four point mutations found in individuals with Sotos syndrome. Open and gray boxes and arrowheads indicate exons, the 5’ and 3’ untranslated regions, and start and stop codons, respectively. Specific domains are indicated by colored boxes, and sequence traces disclose mutations in lower row. c, FISH analysis of the affected individual harboring the deletion. Absence of a FISH signal for RP1-118m12 containing NSD1 (green) along with the presence of 5pter signals (red) on the individual’s chromosome 5 is apparent. d, Summar y of FISH deletions in 20 affected individuals. Known genes, probes used and their genomic locations are indicated in the upper row. Numbers (1-20) and black and blue lines represent affected individuals, regions without deletion and those regions deleted, respectively.

Reprinted with permission: Kurotaki N, et al. Hum Mol Genet. 2005;15:535–542. Copyright © 2005. All rights reserved.

There is a vague phenotype-genotype relationship in Sotos syndrome: thus, macrosomia, developmental delay, and minor anomalies are present in these patients with either intragenic mutations or gene microdeletions. However, patients with cerebral gigantism and microdeletions of NSD1 tend to have rather severe developmental delay and major structural anomalies of the central nervous, cardiovascular, and genitourinary systems, but only modest overgrowth.4,25 On the other hand, patients with intragenic mutations may express less severe anomalies but demonstrate greater linear overgrowth.2,25 Nevertheless, unrelated Sotos syndrome patients with identical mutations in NSD1 may have different phenotypes.4 In neonates with mutations in NSD1, birth length is substantially greater than in subjects with clinical Sotos syndrome but an intact gene.26 Arm span relative to height and hand length relative to age are substantially greater in patients with Sotos syndrome (due to a mutation in NSD1) than in subjects with clinical manifestations of cerebral gigantism and normal NSD1.27 Developmental delay may be more severe in patients with a mutation in NSD1 than in those with clinically diagnosed cerebral gigantism.

Mutations or deletions of NSD1 most often arise de novo and thus the risk of familial recurrence to phenotypically and genotypically normal parents is low. Nevertheless, a patient with Sotos syndrome due to an intragenic mutation has a 50% risk of transmitting this mutated gene to an offspring. 7 No instance of germline NSD1 mosaicism has been observed to date. Intragenic mutations in NSD1 have also been reported in patients with Weaver syndrome (OMIM 277590), an overgrowth disorder that is characterized by macrocrania, hypertelorism, large ears, retrognathia, hypotonia, developmental delay, loose skin folds, dysplastic deeply set nails, sparse hair, and hoarse cry as well as various skeletal anomalies.24 Patients with Weaver syndrome very rarely develop tumors. In 2 out of 52 patients with the Beckwith-Wiedemann syndrome (BWS - OMIM 130650) who presented with in utero macrosomia, macroglossia, hemihyperplasia, and abdominal wall defects, mutations in NSD1 were also reported.28 A microdeletion of NSD1 was detected in an infant girl with features of both Sotos and Nevo syndromes (OMIM 601451).29 However, there is evidence that suggests that the Weaver, BWS, and Nevo syndrome patients studied in these reports were more likely to have had clinical variations of Sotos syndrome.6,30 Anomalies of 11p15, the site of imprinting errors associated with BWS, were identified in 2 out of 20 patients with cerebral gigantism, including one with paternal isodisomy of 11p15 of the H19 locus and one with partial isolated demethylation of KCNQ1OT, perhaps suggesting a functional relationship between NSD1 and the imprinting centers on 11p15.2,28

DIAGNOSIS AND MANAGEMENT

The diagnosis of Sotos syndrome is established by clinical findings (characteristic facial features, prenatal and postnatal overgrowth, persistently enlarged head circumference, developmental delay, and advanced bone age [Table 1]) and confirmed by identification of a mutation in NSD1. Since this disorder is genetically heterogeneous, absence of a mutation does not negate the diagnosis; an abnormal methylation pattern on chromosome 11p15 might be investigated in patients with intact NSD1.28 Cerebral gigantism is to be differentiated from Weaver syndrome and other overgrowth syndromes both on clinical grounds and by the presence of mutations in NSD1. Treatment is symptomatic and focuses on monitoring of growth and periodic surveillance for associated systemic anomalies or development of neoplasms and on factors that may ameliorate developmental delay and behavioral problems. It has been recommended that children with Sotos syndrome be surveyed frequently for tumor development through 10 years of age. Complete physical examinations and blood counts should be done 3 to 4 times each year; abdominal ultrasound studies, α-fetoprotein, and β-hCG measurements twice yearly; and chest x-ray and urine catecholamine determinations once each year (Table 2).10 Given the relatively low incidence of tumors in these patients, these recommendations may be excessive.11

Table 1.

Adapted from reference 12.
Facial characteristics: frontal bossing,high hairline, dolicocephaly, prominent chin, highly arched palate, anti-mongoloid slant of palpebral fissures
Growth - all measurements before adult height
SDS - Standard deviation score
TH - Target height

Table 2.

Adapted from reference 10.

CONCLUSION

Cerebral gigantism is an overgrowth syndrome characterized by increased in utero and postnatal growth, an adult height that is within or slightly above the upper normal range, macrocephaly, a characteristic face, and variable degrees of developmental delay. There is a high incidence of cardiovascular, central nervous and genitourinary malformations in these patients. Tumors occur in 2% to 4% of patients with cerebral gigantism, usually before 8 to 10 years of age. Mutations in NSD1 seem to be quite specific for Sotos syndrome. 6 Indeed, all subjects with documented mutations in NSD1 have some manifestation(s) of cerebral gigantism. The majority of mutations in NSD1 have occurred de novo, but the risk for development of Sotos syndrome in the offspring of a patient with a mutation in NSD1 is 50%. Periodic surveillance for tumor development is recommended in children with Sotos syndrome.

Resources

Sequence analysis of NSD1 may be obtained through The Greenwood Genetics Center, Greenwood, SC.

The Sotos Syndrome Support association may be contacted at www.well.com/user/sssa.

References - (linked to Pubmed Links)

  1. Sotos JF, Dodge PR, Muirhead D, et al. N Engl J Med. 1964;271:109–116.
  2. Cytrynbaum CS, Smith AC, Rubin T, Weksberg R. Curr Opin Pediatr. 2005;17:740–746.
  3. Shaw CJ, Lupski JR. Hum Mol Genet. 2004;(S)13:R57–R64.
  4. Tatton-Brown K, Douglas J, Coleman K, et al. Am J Hum Genet. 2005;77:193–204.
  5. Faravelli F. Am J Med Genet C Semin Med Genet. 2005;137:24–31.
  6. Tatton-Brown K, Rahman N. Clin Dysmorphol. 2004;13:199–204.
  7. van Haelst MM, Hoogeboom JJM, Bruggenwirth HT, et al. Am J Hum Genet A. 2005;139:40–44.
  8. Compton MT, Celentana M, Price B, Furman AC. Psychopathology. 2004;37:190–193.
  9. Mauceri L, Sorge G, Baieli S, Rizzo R, Pavone L, Coleman M. Pediatr Neurol. 2000;22:64–67.
  10. Bouthelier RG, Lapunzina P. J Pediatr Endocrinol Metab. 2005;18(suppl):1227–1235.
  11. Rahman N. Curr Opin Genet Dev. 2005;15:227–233.
  12. de Boer L, Hoogerbrugge CM, van Doorn J, van Buul-Offers SC, Karperien M, Wit JM. J Pediatr Endocrinol Metab. 2004;17:615–627.
  13. Nunn SE, Peehl DM, Cohen P. J Cell Physiol. 1997;171:196–204.
  14. Bunn RC, Fowlkes JL. Trends Endocrinol Metab. 2003;14:176–181.
  15. Rayasam GV, Wendling O, Angrand PO, et al. EMBO J. 2003;22:3153–3163.
  16. Xiao B, Jing C, Wilson JR, et al. Nature. 2003; 421:652–656.
  17. Nielsen AL, Jorgensen P, Lerouge T, Cervino M, Chambon P, Losson R. Mol Cell Biol. 2004;24:5184–5196.
  18. Jaju RJ, Fidler C, Haas OA, et al. Blood. 2001; 98:1264–1267.
  19. Kurotaki N, Imaizumi K, Harada N, et al. Nat Genet. 2002;30:365–366.
  20. Kurotaki N, Stankiewicz P, Wakui K, Niikawa N, Lupski JR. Hum Mol Genet. 2005;15:535–542.
  21. Visser R, Shimokawa O, Harada N, et al. Am J Hum Genet. 2005;76:52–67.
  22. Miyake N, Kurotaki N, Sugawara H, et al. Am J Hum Genet. 2003;72:1331–1337.
  23. Cecconi M, Forzano F, Milani D, et al. Am J Med Genet A. 2005;134:247–24.
  24. Douglas J, Hanks S, Temple IK, et al. Am J Hum Genet. 2003;72:132–143.
  25. Nagai T, Matsumoto N, Kurotaki N, et al. J Med Genet. 2003;40:285–289.
  26. de Boer L, Kant SG, Karpepien M, et al. Horm Res. 2004;62:197–2007.
  27. de Boer L, Le Cessie S, Wit JM. Acta Paediatr. 2005;94:1142–1144.
  28. Baujat G, Rio M, Rossignol S, et al. Am J Hum Genet. 2004;74:715–720.
  29. Kanemoto N, Kanemoto K, Nishimura G, et al. Am J Med Genet. 2006;140:70–73
  30. Turkmen S, Gillessen-Kaesbach G, Meinecke P, et al. Europ J Hum Genet. 2003;11:858–865.

 

 

 


Last Updated: 04/30/2008

Copyright © 2003-2008 Prime Health Consultants, Inc.