Characterization of the Chediak-Higashi syndrome gene in human and mouse

MISSING IMAGE

Material Information

Title:
Characterization of the Chediak-Higashi syndrome gene in human and mouse
Physical Description:
Book
Creator:
Tchernev, Velizar T
Publication Date:

Record Information

Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 29684549
oclc - 51334513
System ID:
AA00020931:00001

Table of Contents
    Title Page
        Page i
        Page ii
    Dedication
        Page iii
    Acknowledgement
        Page iv
    Table of Contents
        Page v
        Page vi
    Abstract
        Page vii
        Page viii
    Introduction
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
    Materials and methods
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
    Results
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
    Discussion
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
        Page 117
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
    Appendix. Sequences of molecules interacting with lyst and lyst2
        Page 123
        Page 124
        Page 125
        Page 126
        Page 127
        Page 128
        Page 129
        Page 130
        Page 131
        Page 132
        Page 133
        Page 134
        Page 135
        Page 136
        Page 137
        Page 138
        Page 139
        Page 140
        Page 141
        Page 142
        Page 143
        Page 144
        Page 145
        Page 146
        Page 147
        Page 148
        Page 149
        Page 150
        Page 151
        Page 152
        Page 153
        Page 154
        Page 155
        Page 156
        Page 157
        Page 158
        Page 159
        Page 160
        Page 161
        Page 162
        Page 163
        Page 164
        Page 165
        Page 166
        Page 167
        Page 168
        Page 169
        Page 170
        Page 171
        Page 172
        Page 173
        Page 174
        Page 175
        Page 176
    Bibliography
        Page 177
        Page 178
        Page 179
        Page 180
        Page 181
        Page 182
        Page 183
        Page 184
        Page 185
        Page 186
        Page 187
        Page 188
        Page 189
        Page 190
        Page 191
    Biographical sketch
        Page 192
        Page 193
        Page 194
        Page 195
Full Text











CHARACTERIZATION OF THE CHEDIAK-HIGASHI SYNDROME GENE IN
HUMAN AND MOUSE















By

VELIZAR T. TCHERNEV


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


1998





























Copyright 1998

by

Velizar T. Tchemev

































To my parents














ACKNOWLEDGMENTS


The author would like to thank his mentor, Dr. Stephen F. Kingsmore, and the

supervisory committee members, Dr. M. Wallace, Dr. E. Wakeland, Dr. E. Sobel and Dr.

M. Bubb, for their valuable advice and encouragement.

The author's appreciation is extended to all former and present members of Dr.

Kingsmore's laboratory, Dr. Maria D.F.S. Barbosa, Dr. Vishnu S. Mishra, John C. Detter,

Dr. Elizabeth B. McMurtrie, Quan A. Nguyen, Sandra M. Holt, Juan M. Teodoro and

Andrea Hofig, as well as to Dr. Krishnan Nandabalan and Dr. Madan Kumar from

CuraGen Corporation, who contributed to the successful completion of this work.

The author wishes to express his gratitude to Ms. Anna V. Gueorguieva and Mr.

Nathan S. Collier for their kindness and continuous support.

Finally, but most importantly, the author would like to thank his parents Lilia

Jeleva and Tzvetan Tchemev for their loving care and guidance, his beloved wife Ralitza

Gueorguieva for her invaluable support and inspiration, and his fellow-players Ivajlo

Kortezov, Rada Kortezova, Ludwig Lubih, Anastas Assenov and Isak Beraha for their

true friendship and thought-provoking conversations.














TABLE OF CONTENTS
page


ACKNOWLEDGMENTS ................................................................................................. iv

A B ST R A C T ...................................................................................................................... vii

IN TR O D U C TIO N ............................................................................................................... 1

MATERIALS AND METHODS....................................................................................... 22
Reverse Transcription-Polymerase Chain Reaction (RT-PCR)
Amplification of CHS Patient and Beige Mouse cDNAs...................................... 22
TA Cloning of PCR Products and Plasmid Purification.............................................. 23
Single-Strand Conformation Polymorphism (SSCP) Analysis ................................... 24
DNA Sequencing and Sequence Analysis ................................................................... 24
Allele-Specific Oligonucleotide Analysis.................................................................... 25
PCR Amplification of Mouse Lyst cDNA Isoforms .................................................... 25
Cloning of Human LYSTcDNAs................................................................................. 26
Molecular Probes ......................................................................................................... 28
Isolation of LYST2 cDNA Clones................................................................................ 29
Southern Blot Analysis and Autoradiography ............................................................. 30
Northern Blot Analysis................................................................................................ 30
Simple Sequence Length Polymorphism (SSLP) PCR Amplification........................ 31
Backcross Mouse Panel and Genetic Mapping............................................................ 31
Cloning of LYST and LYST2 Bait cDNA Fragments in Yeast Two-
H ybrid V ectors....................................................................................................... 32
Yeast Two-Hybrid Screens.......................................................................................... 36
Confirmation of the Specificity of Protein Interactions by the Yeast
Two-Hybrid System............................................................................................... 40

R E SU L T S .......................................................................................................................... 43
Identification of Mutations in Patients with Chediak-Higashi
Syndrome and in Beige Mice................................................................................. 43
Identification and Characterization of Lyst mRNA Isoforms...................................... 52
Identification and Characterization ofLyst2, a Brain-Specific Member
of the Chediak-Higashi Syndrome Gene Family................................................... 61
Identification and Sequence Analysis.................................................................... 61
Genetic Mapping.................................................................................................... 62
Expression Analysis............................................................................................... 67








Identification of Proteins that Interact with the CHS Protein and with
LYST2, Using a Yeast Two-Hybrid Approach ..................................................... 71
IP-1 (EST cg50136.f6, APPENDIX, Sequence 13)............................................... 76
IP-2 (EST AA010799, APPENDIX,.Sequence 14)............................................... 77
IP-3 (EST cg50136.c10, APPENDIX, Sequence 15)............................................ 77
IP-4 (EST cg50136.a7, APPENDIX, Sequencel6) ............................................... 78
IP-5 (EST cg50175.c7, APPENDIX, Sequence 17) .............................................. 79
IP-6 (EST cg50138.g5, APPENDIX, Sequence 18).............................................. 80
IP-7 (EST cg50173.dlO, APPENDIX, Sequence 19)............................................ 81
IP-8 (EST cg50175.h7, APPENDIX, Sequence 20).............................................. 82
IP-9 (EST KIAA0192, APPENDIX, Sequence 21)............................................... 83
IP-10 (EST cg50136.a5.b, APPENDIX, Sequence 22) ......................................... 83
IP-11 (EST cg51287.dl 0, APPENDIX, Sequence 23).......................................... 84
IP-12 (EST cg49432.h3.b, APPENDIX, Sequence 24)......................................... 85
IP-13 (EST cg50175.c ll, APPENDIX, Sequence 25).......................................... 86

D ISC U SSIO N .................................................................................................................... 88
Identification and Characterization of Lyst mRNA Isoforms ...................................... 88
Identification of Mutations in Patients with Chediak-Higashi
Syndrom e and in Beige M ice................................................................................. 92
Identification and Characterization of Lyst2, a Brain-Specific Member
of the Chediak-Higashi Syndrome Gene Family................................................... 94
Identification of Proteins that Interact with the CHS Protein, Using a
Yeast Two-Hybrid Approach................................................................................. 98

APPENDIX SEQUENCES OF MOLECULES INTERACTING WITH LYST
A N D L Y ST 2 .............................................................................................................. 123

B IB LIO G R A PH Y ............................................................................................................ 177

BIOGRA PHICAL SKETCH ........................................................................................... 192














Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

CHARACTERIZATION OF THE CHEDIAK-HIGASHI SYNDROME GENE IN
HUMAN AND MOUSE

By

Velizar T. Tchemrnev

December 1998

Chairman: Stephen F. Kingsmore
Major Department: Pathology, Immunology and Laboratory Medicine

Chediak-Higashi syndrome (CHS) is an autosomal recessive, immune deficiency

disorder of human and mouse (beige, bg) that is characterized by abnormal intracellular

protein transport to and from the lysosome/late endosome. Recent reports have described

the positional cloning of homologous genes, LYST and Lyst, that are mutated in human

CHS and in bg mice, respectively. However, since the encoded proteins were novel and

unlike any of the molecules previously implicated in vesicular trafficking, their

identification did not prove immediately helpful in establishing the precise mechanism

whereby CHS dysregulates protein transport. Therefore, the main goal of this dissertation

was to characterize the Chediak-Higashi gene and its products in more detail using

functional genomics approaches. We demonstrated that each of the previously reported

bg gene sequences are derived from a single gene with alternatively spliced mRNAs.

Alternative splicing results in bg gene isoforms (Lyst-I and Lyst-II) that contain different









3' regions. Similarly to mouse, the human mRNA isoforms arise from incomplete

splicing and retention of a transcribed intron that encodes the C-terminus of the predicted

LYST protein. Additional splicing complexity of smaller isoforms exists. The Lyst-Il

isoform lacks exons a and 13, while in Lyst-IV, exons a, 13 and y are absent. Novel

mutations were identified within the coding domain of LYST in several CHS patients and

bg alleles. Interestingly, all bg and LYST mutations identified to date are predicted to

result in either truncated or absent proteins. Mutation and expression analyses suggested

that defects in the full-length mRNA alone can elicit Chediak-Higashi syndrome and that

expression of the smaller isoform alone cannot compensate for loss of the largest isoform.

We have also identified a novel gene in human (LYST2) and mouse (Lyst2) that appears

to be a relative of the CHS gene, based on sequence similarity, predicted protein structure

and on similar transcript size. Comparison of the relative abundance of LYST2 and LYST

mRNAs suggests that LYST2 is expressed abundantly only in brain and therefore it may

represent a brain-specific member of the CHS gene family. Using a modified yeast two-

hybrid system, a human cDNA library was screened with baits from the coding domains

of LYST and LYST2. Several proteins, which play important roles in protein transport and

signal transduction, such as 14-3-3, casein kinase II, calmodulin and Hrs, were found to

interact with LYST and/or LYST2. Many of the interacting proteins could be linked in a

common pathway that regulates vesicular trafficking and degranulation.














INTRODUCTION


Chediak-Higashi syndrome (CHS) is a primary immunodeficiency, characterized

by severe recurrent infections (Blume & Wolff, 1972), partial albinism (Windhorst et al.,

1968), nervous system abnormalities (Misra et al., 1991), high incidence of malignancy

(Hayakawa et al., 1986) and predisposition to bleeding (Meyers et al., 1974). Eighty-five

percent of patients undergo an accelerated lymphoma-like phase, characterized by fever,

hepatosplenomegaly, lymphadenopathy, pancytopenia, coagulopathy and widespread

lymphohistiocytic organ infiltrates. This complication is fatal in the absence of bone

marrow transplantation.

The first 3 cases of Chediak-Higashi syndrome were reported by Beguez-Cesar in

1943, then one case by Steinbrink in 1948, and one case by Higashi in 1954. Chediak

published certain aspects of Beguez-Cesar's cases in 1952, and Sato in 1955 recognized

the similarities between Chediak's and Higashi's data, inventing the eponym "Chediak-

Higashi syndrome." Except for the occasional usage of single names or combinations of

the above authors' names, the disease has been most commonly referred to as the

Chediak-Higashi syndrome.

Approximately 150 cases of CHS have been reported in the literature. Males and

females are affected equally. The mean age of death of CHS patients is 6 years. The

Chediak-Higashi syndrome occurs with a low frequency throughout the world. Europe,






2


America and Japan are the areas from which the majority of the known cases have been

described. While most of the reports deal with isolated cases, a unique cluster of Chediak-

Higashi syndrome patients has also been published (Ramirez-Duque et al., 1982).

Fourteen cases of CHS were found in 12 families living in a defined and relatively

isolated geographical area (Pregonero) in the Venezuelan Andes. The patients were pre-

school children except one 11-year-old female. Six of the patients were male. All showed

the same typical clinical characteristics of the syndrome. Since CHS is an autosomal

recessive disease (Sato, 1955), the existence of this high-frequency cluster of CHS

patients in a restricted region could be explained by a founder effect in an isolated

population with inbreeding.

The presence of giant, perinuclear vesicles with protein markers characteristic of

late endosomes and mature lysosomes is pathognomonic for CHS (Burkhardt et al.,

1993). The organelles are abnormal both in size (enlarged) and in cytoplasmic

distribution (perinuclear). While the late endosomal/lysosomal compartment is uniformly

affected in CHS, the histological designation of affected organelles varies by cell type:

melanosomes are affected in CHS melanocytes, while secretary granules are affected in

CHS neutrophils, eosinophils, mast cells, cytolytic T-cells, proximal renal tubules, and

lysosomes are affected in monocytes and fibroblasts. The giant organelles are thought to

result from dysregulated fusion of primary endosomes and lysosomes (Burkhardt et al.,

1993; Willingham et al., 1981). The morphological abnormalities of bg and CHS

lysosomes are accompanied by compartmental mis-sorting of proteins such as elastase, 3-

glucuronidase, and cathepsin G (Holcombe et al., 1994).









The disease also occurs in other mammalian species mouse, rat, cat, cattle,

mink, fox and killer whale. Beige (bg) is a well described mouse model of CHS. The

clinical and pathological features of CHS patients and bg mice are indistinguishable.

Prior to the cloning of bg, the strongest evidence of homologous defects in beige and

CHS came from interspecific genetic complementation studies, which demonstrated that

fusion of bg and CHS fibroblasts failed to reverse lysosomal abnormalities, while fusion

with normal cells did eliminate cytological defects of bg or CHS fibroblasts (Perou and

Kaplan, 1993; Penner and Prieur, 1987).

Partial albinism is one of the main clinical manifestations of the Chediak-Higashi

syndrome. Most patients have hypopigmentation of their hair, skin and ocular fundi, as

well as misrouting of the optic nerve fibers that is characteristic of all types of albinism.

Windhorst et al., 1966, 1968 reported the presence of giant melanosomes in melanocytes

from CHS patients. Some melanosomes were highly irregular, with a suggestion of fusion

of separate granules. Most epidermal melanosomes from a CHS patient, observed by

Zhao et al., 1994, were several times larger than normal, relatively amelanotic, and

occasionally clustered into complexes. No melanosomes were detected in keratinocytes

adjacent to melanocytes, suggesting that the normal transport of melanosomes from

melanocytes to neighboring keratinocytes is defective. After DOPA histochemistry, used

to localize functional tyrosinase, reaction product was observed throughout the

cytoplasm, but not in most of the aberrant large melanosomes and complexes, indicating

that functional tyrosinase is either not translocated to or is not retained within the giant

granules. In contrast to control melanocytes, DOPA-positive vesicles migrated far beyond

their usual perinuclear and trans Golgi network location and were seen in the vicinity of









the plasma membrane in melanocytes from CHS patients. When several melanocyte-

specific proteins were localized histochemically or immunohistochemically, they were

predominantly detected in the perinuclear area and less prevalent in the dendrites of CHS

cells, compared to their relatively uniform distribution throughout the cytoplasm of

control cells (Zhao et al., 1994), suggesting an inefficient intracellular translocation of

these proteins. Medium, conditioned by CHS cells for 1 to 3 days, displayed increasing

amounts of tyrosinase activity that did not exist in medium conditioned by numerous

control cell lines. P3-glucuronidase was also released into the culture medium by CHS

melanocytes but not by control melanocytes. These data suggest that the low tyrosinase

activity in giant melanosomes results from mislocalization and eventual extracellular

secretion of tyrosinase. This hypothesis is supported by the presence of DOPA-positive

vesicles beneath the plasma membrane where fusion/secretion could occur.

Patients with Chediak-Higashi syndrome, especially those who survive into

adulthood, and elderly beige mice, develop serious neurologic defects. Lockman et al.,

1967, reported a patient with the typical clinical and cytological picture of CHS and a

severe motor neuropathy. A median nerve biopsy showed that many Schwann cells

contained large cytoplasmic particles which the authors considered to be abnormal

lysosomes. Two cases with lymphohistiocytic infiltration of the sural and femoral nerves

were described by Sung et al., 1969. The myelin sheaths and axons showed swelling and

fragmentation or had disappeared. Schwann cells were hypertrophic and contained

cytoplasmic inclusions. Ultrastructural examination of dorsal root and sympathetic

ganglion cells demonstrated inclusions resembling lysosomes and lipofuscin granules.









Pettit and Berdal, 1984, reported a 25-year-old man with a neurologic disorder that

resembled a spinocerebellar degeneration and Parkinsonism. Cerebellar cortical atrophy

(Kondo et al., 1994) and diffuse atrophy of the brain with abnormalities in the

periventricular and corona radiata regions were detected by computer tomography and

magnetic resonance imaging (Ballard et al., 1994). The oldest known CHS patient first

seen with a neurologic disorder in early adult life was described by Uyama et al., 1994.

This 39-year-old woman developed mental deterioration, Parkinsonism, muscular atrophy

of limbs, and loss of tendon reflexes. MRI showed marked temporal dominant brain

atrophy and diffuse spinal cord atrophy. Similarly, beige mice that survive to 17 months

of age show a progressive neurologic disorder accompanied by nearly complete loss of

cerebellar Purkinje cells (Murphy and Roths, 1978). Guo et al., 1992, reported

cytoarchitectonic abnormalities in the brains of beige mice. In the cerebellum, Purkinje

cells and clusters of granule cells, and occasionally glia cells, were located ectopically in

the molecular level. In the hippocampus, ectopically situated pyramidal cells were found.

The infrapyramidal mossy fiber layer was not compact, but appeared as clumps of

scattered fiber bundles. Pezeshkpour et al., 1986, reported the presence of giant

lysosomes in Schwann cells in a nerve biopsy from a CHS patient. Peripheral neuropathy

in a 33-year-old brother and a 29-year-old sister with CHS were described by Misra et al.,

1991. Both showed evidence of a sensory neuropathy associated with central nervous

system involvement. Nerve conduction studies indicated an axonal neuropathy. Sural

nerve biopsy demonstrated a loss of myelinated nerve fibers, particularly those of larger

size, and of unmyelilnated axons. By light and electron microscopy, granulated cells with

the staining properties of mast cells were present in the endoneurium. Instead of being the









usual 0.1-0.2 pm in diameter, the granules were approximately 10 times larger and with

unusual appearance. Blume and Wolf, 1972, reported irregular giant inclusions, thought

to be lysosomes, in the cytoplasm of Schwann cells. These morphologic observations of

abnormally large granules suggest that the neurological aberrations in CHS result from

defective vesicular transport of molecules that are important for the normal cellular

function.

Bleeding tendency is another major clinical manifestation in patients and animals

with CHS. While the platelet numbers are normal (Novak et al., 1985; Phillips et al.,

1967), dense platelet granules are virtually absent, whole-blood serotonin is markedly

decreased, and platelets have reduced ability to accumulate serotonin (Meyers et al.,

1974; Meyers et al., 1983). The prolonged bleeding time and low serotonin

concentrations were converted to normal values following transplantation of normal bone

marrow in beige mice. Likewise, control mice displayed symptoms of storage pool

deficiency when transplanted with mutant marrow (Novak et al., 1985). These studies

demonstrate that the CHS platelet storage pool deficiency results from a defect in bone

marrow precursor cells. In agreement with this conclusion, Menard and Meyers, 1988,

showed that the dense granule precursors in maturing and mature megakaryocytes were

absent. Abnormal platelet aggregation was described in 6 patents with CHS (Apitz-Castro

et al., 1985). The total content and the maximal amounts of the dense granule constituents

secreted after thrombin stimulation were greatly decreased. CHS platelets loaded with

radiolabelled serotonin showed a spontaneous release of radioactivity not observed in

normal platelets under the same conditions. Similar defects were reported in cats (Colgan









et al., 1989) and in blue foxes (Sjaastad et al., 1990) with Chediak-Higashi syndrome. In

beige mice, platelet aggregation after stimulation was significantly decreased compared

to that of normal mice (Pratt et al., 1991). Platelets from beige mice were approximately

10 times more sensitive to prostacyclin inhibition of collagen-induced aggregation than

control platelets. The stores of serotonin and adenine nucleotides were decreased. Rendu

et al., 1983, reported a large reduction in the number of serotonin-storage granules (dense

bodies) but otherwise normal ultrastructure and normal amounts of a- and catalase-

positive granules in 3 patients with Chediak-Higashi syndrome. The platelet release

reaction, studied with thrombin as an inducer, was impaired. The serotonin uptake by the

patients' platelets was low and its metabolism was increased. These findings show that

human CHS platelets are deficient in the storage pool of dense granule substances and

suggest that this granule defect influences the release mechanism of other granule

constituents, leading to abnormal aggregation and bleeding predisposition.

Most morbidity and mortality in CHS result from immune deficiency, with severe

defects in innate immunity granulocytess and NK cells) and relatively mild impairment of

acquired immune function (B and T cells).

The initial immune defect recognized in CHS was a striking reduction in the

ability of granulocytes to kill phagocytosed microorganisms (Blume et al., 1968; Gallin

et al., 1974). Phagocytosis of radiolabelled Staphylococcus aureus by CHS granulocytes

was normal. However, there was reduced bactericidal activity of S. aureus and group D

Streptococcus by CHS cells through 90 min of incubation. This defect was most

pronounced at early time periods and was related to impaired intracellular killing. There









are occasional reports suggesting more widespread neutrophil abnormalities. In one study

(Bellinati-Pires et al., 1992), in which phagocytosis and killing of Staphylococcus aureus

were simultaneously evaluated by a fluorochrome phagocytosis assay, both appeared to

be deficient in neutrophils form two patients with CHS. The presence of giant

phagolysosomes, enclosing bacteria in active proliferation 45 min after phagocytosis was

noteworthy, and corresponded with the impaired bactericidal activity of these leukocytes.

In an under-agarose assay, phagocytes of homozygote and phenotypically normal

heterozygote CHS cats recognized and responded equally well to bacterial stimuli as did

cells from control animals but traveled shorter distances primarily because of a reduced

inherent motility. Similar results were obtained when phagocyte chemotaxis was

evaluated with zymosan-activated serum (Colgan et al., 1992). Mildly defective CHS

neutrophil and monocyte locomotion was discovered by the micropore filter technique

using modified Boyden chambers (Yegin et al., 1983). Isolated reports suggest that

diminished surface expression of C3bi receptor and decreased oxygen radical generation

by polymorphonuclear leukocytes (PMNs) might be some of the factors to explain the

moderately defective PMN mobility, chemotaxis and bactericidal activity (Cairo et al.,

1988; Kubo et al., 1987). While the literature contains some conflicting data, the

consensus is that, with the exception of the ability to kill phagocytosed microorganisms,

other granulocyte activities, such as phagocytosis, chemotaxis and protein degradation,

are relatively well preserved in CHS (Gallin et al., 1974; Clawson et al., 1978).

Analogous defects occur in CHS monocytes and macrophages -- cytolysis of tumor

targets is impaired, while phagocytosis and chemotaxis are almost normal (Gallin et al.,

1975; Mahoney et al., 1980). CHS selectively involves crystalloid granules in late









eosinophils, while the non-crystalloid granules are preserved, suggesting that the

underlying pathology does not affect all lysosomal subpopulations (Hamanaka et al.,

1993).

The second category of CHS immune defects is a virtual ablation of natural killer

(NK) cell activity and antibody dependent cellular cytotoxicity (Roder and Duwe, 1979;

Roder et al., 1979, 1982). Mature NK cells demonstrate low lytic efficiency, assessed by

conventional 4 h chromium release assay and by a more recently developed single cell

liquid cytotoxic assay. The NK cells were not activated by prolonged in vitro incubation

with IFN-y, but they exhibited IL-2 induced cytotoxicity, though the magnitude of

induction was uniformly less than in controls (Virelizier and Griscelli, 1980; Holcombe,

1992). CHS NK cells have the necessary cellular structures required for their role as lytic

effector cells, but lack cytotoxic function due to a relative refractoriness in initiating the

post-binding lytic mechanism. NK cells in CHS patients are blocked at a post-

recognition, post-activation step in the cytolytic pathway, subsequent to the burst of

oxygen intermediates but preceding the lethal hit. In summary, while NK cells are present

in normal numbers, bind target cells effectively and generate superoxide ions normally,

exocytosis of lytic granules from the CHS NK cells is refractory (Targan and Oseas,

1983; Roder et al., 1983; Brahmi, 1983).

Tumor susceptibility of bg mice was examined as a direct test of the hypothesis

that NK cells are involved in surveillance against neoplasia. Subcutaneous inoculation in

a transplantation test yielded higher tumor take incidence in bg/bg mice than in

heterozygous littermates. The differences were most striking in the early phase of the









observation period, with a majority of the total tumor takes in bg/bg registered within 2

weeks, whereas only a small part of the tumors in bg/+ mice were palpable at this time.

Thus, in addition to the increased incidence in homozygous mice, the progressively

growing tumors appeared with a shorter latency than in controls. This was also reflected

in the larger mean tumor diameter by 2 and 3 weeks and the earlier deaths among tumor-

bearing bg/bg than bg/+ mice. These findings correlated well with the in vitro measured

splenic NK activity against both tumors tested. Tumor cells had a low, but significant,

sensitivity to bg/+ spleen cell cytotoxicity (5-10 % specific lysis), whereas bg/bg spleen

consistently gave values below 3% specific lysis (Kaerre et al., 1980). In another study, a

tumor line modified to be sensitive to NK cytotoxicity in vitro, demonstrated in vivo an

increased growth rate, faster induction time and an enhanced metastatic capability in bg

compared to control mice. This was not found with a tumor line insensitive to NK

activity. In vivo activation of NK cells in bg and control mice resulted in decrease in

tumor growth rate and metastatic frequency (Talmage et al., 1980). In a large Japanese

cohort, non-Hodgkin's lymphoma occurred in over one third of CHS patients (Hayakawa

et al., 1986). These results indicate that NK cells have an important function in the host's

control of tumor growth and metastasis. The idea to use the bg model to dissect the

contribution of NK cells to immunity, and in particular, to tumor immunosurveillance

was dispelled somewhat, however, by the subsequent recognition of concomitant

impairment, albeit mild, in T and B cell function.

Although both B and T cells are affected in CHS, the abnormalities in the specific

defense mechanisms are not as severe as those in natural immunity and have less

importance for the pathogenesis of CHS. The pathognomonic giant cytoplasmic vesicles









are detected in B cells from CHS patients following activation and differentiation (Grossi

et al., 1985). Moderate and variable reduction of LPS-induced B cell mitogenesis is

demonstrated in CHS. The in vivo anti-trinitrophenyl antibody response to a thymus-

independent antigen (TNP-Ficoll) was found to be significantly lower in C57BL/6J-bg

than in C57BL/6J mice, although both strains responded similarly to an analogous

thymus-dependent antigen (TNP-ovalbumin) (Pflumio et al., 1990). Since the marginal

zone macrophages of the spleen were previously shown to be essential for the initiation of

antibody responses to thymus-independent antigens, they might be another target of the

bg mutation, possibly resulting in impaired antigen presentation to B cells.

It is notable that homozygosity at the bg locus confers protection against immune

complex glomerulonephritis to BXSB mice with systemic autoimmune disease (Clark et

al., 1982). However, this phenomenon is probably not due to a decreased B cell function

because the levels and types of autoantibodies remain unchanged in bg homozygotes.

Extremely high antibody titers to the Epstein-Barr virus specific viral capsid

antigen, to the restricted component of the EBV-induced early antigen complex and to the

EBV-associated nuclear antigen were detected in CHS patients (Merino et al., 1983). It is

likely that the elevated antibody titers reflect an increased production of these antigens

due to defective NK and antibody-dependent cellular cytotoxicity (ADCC) activities.

Cytotoxic lymphocytes (CTLs) from CHS patients are unable to destroy target

cells recognized via the T cell receptor (TCR). Individual CTL clones show poor killing

that can be increased in longer assays. However, in the presence of cycloheximide, the

small amount of killing is abolished, indicating that killing arises from newly synthesized

proteins, rather than from proteins stored in granules. It has been shown that the CHS









CTL clones express normal levels of the lytic proteins granzyme A, granzyme B and

performin, which are processed properly during biosynthesis and targeted correctly to giant

lytic granules. Despite the difference in size, CHS and normal lytic granules are similar,

both contain the lysosomal enzyme cathepsin D and granzyme A, and lack the mannose-

6-phosphate receptor. However, unlike normal CTL clones, the CHS CTLs are unable to

secrete their giant granules in which the lytic proteins are stored. After cross-linking the

TCR, CHS CTL clones fail to secrete granzyme A, as assayed by both enzyme release

and confocal microscopy (Baetz et al., 1995).

The proportion of gamma-delta T cells among peripheral blood mononuclear cells

is significantly increased in CHS. The cellular machinery for lysis of target cells in vitro

is present in CHS-derived gamma-delta T cell clones (Holcombe et al., 1990). This is the

first example of a specific immunodeficiency disorder with a relative expansion of these

T cells.

A short summary of the immunological features of CHS is presented in Table 1.











Table 1. Main immunologic features in Chediak-Higashi syndrome


INNATE IMMUNITY ACQUIRED IMMUNITY



Granulocytes NK cells B cells T cells



1. Greatly reduced 1. Very low lytic 1. Reduced antibody 1. Decreased ability

ability to kill activity caused by response to thymus- of CTLs to kill target

phagocytosed micro- defective exocytosis independent antigens cells, caused by

organisms of lytic granules (may be due to defective secretion of

2. Chemotaxis, 2. Numbers of NK impaired natural lytic granules

phagocytosis and cells, binding to tar- immunity) 2. Relative expansion

other activities are get cells, and oxygen 2. Mildly reduced of gamma-delta T

relatively well radical production are proliferative response cell population

preserved normal to LPS





PATHOGNOMONIC GIANT PERINUCLEAR VESICLES








Most mortality in CHS patients is caused by two manifestations of their immune

deficiency -- severe pyogenic infections and the accelerated phase. Lung, upper

respiratory tract and skin infections caused by Staphylococcus, Streptococcus and

Pneumococcus predominate in CHS, although a wide variety of infection sites and

organisms occur. The accelerated phase is considered a clinicopathologic expression of a

virus-associated lymphoproliferation because of low CD4' to CD8+ T lymphocyte ratio,

presence of Epstein-Barr virus genome in the mononuclear cells of the lymph node, blood

and bone marrow, and possible clinical responses to acyclovir. The current treatment for

CHS patients ranges from vitamins to cytostatic agents and bone marrow transplantation,

but the prognosis remains poor because of increasing susceptibility to infections and

progressive neurologic deterioration. Several studies suggest novel strategies for

immunomodulation: recombinant canine granulocyte colony-stimulating factor to

improve the neutrophil function (Colgan et al., 1992), IL-2 to induce cytotoxicity of CHS

lymphocytes (Holcombe, 1992), and high-dose intravenous gammaglobulin for

management of the accelerated phase (Kinugawa and Ohtani, 1985). The goal of such

therapies would be to partially compensate the immune cell dysfunction caused by the

basic genetic defect and manifested as abnormal protein trafficking.

The transport of newly synthesized proteins from the endoplasmic reticulum to

the Golgi apparatus and from there to the cell surface and elsewhere, is mediated by

transport vesicles, which transfer enclosed proteins by cycles of vesicle budding and

fusion. The abnormalities in CHS can be explained by a compartmental mis-sorting of a

variety of proteins, which are necessary for the proper function of the affected cell types,

for example CHS CTLs cannot secrete stored granzyme A and other lytic proteins, and









therefore cannot lyse target cells efficiently. The morphologic manifestation of the

protein sorting defect in CHS is the presence of giant perinuclear vesicles in the

cytoplasm of the affected cells. The assortment of protein trafficking defects observed in

CHS defies simple explanation: trafficking both to and from the late

endosomal/lysosomal compartments is affected, and proteins which utilize both the

cation-independent mannose-6-phosphate and other transport mechanisms are mis-sorted.

In contrast, targeting of membrane proteins to lysosomal/late endosomal compartments

(LAMP1, LAMP2, rab7, acid phosphatase) is intact. Experiments demonstrating

spontaneous concavalin-A induced capping of CHS granulocytes suggest that the defect

impairs interactions between microtubules and endosomes/lysosomes (Oliver et al.,

1975). However, detailed examination of microtubules and microtubule-based motors has

not disclosed an abnormality (Perou and Kaplan, 1993). In summary, abnormalities in

lysosomal and endosomal processing may explain the clinical features of CHS,

suggesting that the defect involves a protein that regulates certain aspects of trafficking to

and from lysosomal/late endosomal compartments in many cell types.

A summary of the mechanisms involved in the pathogenesis of Chediak-Higashi

syndrome is presented in Fig. 1. All clinical manifestations are caused by a common

defect in the regulation of vesicular trafficking in different cell types: melanosomes are

affected in melanocytes, resulting in partial albinism; dense granules in platelets are

defective, leading to bleeding predisposition; lysosomes and secretary vesicles in

numerous immune cells are deficient, causing frequent infections and increased incidence

of tumors; and giant inclusions are present in Schwann and Purkinje cells, which could be

associated with neurological abnormalities.






BASIC MOLECULAR DEFECT


ABNORMAL LYSOSOMAL/LATE ENDOSOMAL COMPARTMENT (GIANT VESICLES)


MELANOSOMES



MELANOCYTES


LYSOSOMES



MONOCYTES

MACROPHAGES

NEUTROPHILS


SECRETARY VESICLES



EOSINOPHILS

MAST CELLS

NK CELLS

LYMPHOCYTES


DENSE GRANULES



PLATELETS


GIANT INCLUSIONS



SCHWANN CELLS

PURKINJE CELLS


$

PARTIAL

ALBINISM


FREQUENT SEVERE INFECTIONS

INCREASED TUMOR INCIDENCE


BLEEDING

PREDISPOSITION


NEUROLOGICAL

ABNORMALITIES


Figure 1. Mechanisms involved in the pathogenesis of Chediak-Higashi syndrome.









The need for molecular understanding of the basic defect in Chediak-Higashi

syndrome led to the first mapping studies, which were greatly facilitated by the existence

of the mouse model beige. Linkage of satin (sa) with bg defined a new linkage group,

later found to lie on proximal mouse chromosome (Chr) 13 (St. Amand and Cupp, 1958;

Lyon and Meredith, 1969; Lane, 1971). Owen et al., 1986, showed that in mouse, bg is

closely linked to the gene for the y-chain of the T cell receptor (TCR). However, in 3

families with CHS, RFLPs in the TCR y-gene were inherited discordantly with CHS,

demonstrating non-linkage (Holcombe et al., 1987) and suggesting a break in the region

of homology between human chromosome 7p and mouse chromosome 13. Further

refinement of the genetic map of proximal mouse Chr 13 (Justice et al., 1990) led to

linkage of bg to the gene encoding nidogen (Nid), a structural basement membrane

protein (Jenkins et al., 1991). No crossovers between Nid and bg were detected in 123

mice, suggesting that the two loci are within approximately 2.4 cM. Since human nidogen

(NID) has been assigned to Chr lq43 (Olsen et al., 1989), the close linkage between bg

and Nid established a new group of homology between mouse Chr 13 and human Chr 1 q,

suggesting that the gene responsible for the Chediak-Higashi syndrome is located in the

telomeric region of human chromosome I q. These results confirmed the earlier

observation of non-linkage between CHS and TCR-y (Holcombe et al., 1987) and the

existence of a breaking point in the region of homology between mouse Chr 13 and

human Chr 7p. Indeed, linkage between NID and CHS on human chromosome lq was

reported by Goodrich and Holcombe in 1995. Homozygosity mapping using markers

derived from distal human chromosome I q was carried out by Fukai et al., 1996, in four









inbred families or probands with typical childhood Chediak-Higashi syndrome. The

human CHS gene was localized to an 18.8 cM interval, flanked by D1S446 and D1S184,

on Chr lq42-q44. Barrat et al., 1996, studied 10 CHS patients from 9 families and

mapped the CHS locus to an approximately 5 cM region between D 1S163 and D1 S2680

on human chromosome lq42.1-q42.2. As a first step in a positional cloning effort,

Kingsmore et al., 1996a, precisely localized the beige gene to a 0.24 cM interval on

proximal mouse Chr 13 by segregation analysis. In a total of 726 mice from 3 different

backcrosses, bg was found to cosegregate with 6 genetic markers (Nid, Estm9, D13Mit56,

D13Mit]162, D13Mit237 and D13Mit240). The candidacy of Nid and Estm9 was

evaluated by Southern, northern and RT-PCR analyses, which suggested that neither of

these two genes represented the bg gene. However, interstrain differences in pulsed field

restriction fragment lengths were observed, providing a useful and simple test for initial

evaluation of candidate genes. A refined physical map of the bg critical region was

reported by Kingsmore et al., 1996b. Interspersed repetitive element-polymerase chain

reaction (IRE-PCR) and direct cDNA selection were used to identify 20 novel sequence

tag sites which enabled the construction of 2 contigs composed of overlapping YAC and

P 1 clones, covering approximately 2400 kb of the bg non-recombinant interval. In 1996,

two groups simultaneously reported the identification of the homologous beige and

Chediak-Higashi syndrome genes. Barbosa et al., 1996, isolated a gene designated Lyst

(Lysosomal trafficking regulator), from YAC 195A8 by direct selection with mouse

spleen cDNA. It was shown that Lyst represented the beige gene with a complete open

reading frame of 4635 nucleotides, encoding a protein of 1545 amino acids. It was also

demonstrated that corresponding human clones with high degree of homology to mouse









Lyst represented the gene responsible for Chediak-Higashi syndrome. Perou et al., 1996b,

also reported the identification of the beige gene by YAC complementation and

positional cloning. Interestingly, the partial cDNA sequence isolated by Perou et al., did

not match or even overlap with the sequence of Lyst, described by Barbosa et al. One of

the aims of the present dissertation was to explain this discrepancy. This issue was

resolved by the demonstration that each of the previously reported beige gene sequences

was derived from a single gene by alternative splicing (Barbosa et al., 1997; Nagle et al.,

1996), as described in the RESULTS section of this dissertation. However, the

expectations that the identification of the complete mouse and human cDNA sequences of

the CHS gene would elucidate the precise function of the encoded protein, were not met.

The 11403-bp full-length open reading frame corresponded to a polypeptide of 3,801

amino acids, with an estimated molecular mass of 429,153 daltons. Although several

putative domains were found in Lyst, the predicted protein was novel and unlike any of

the molecules previously implicated in vesicular transport (Barbosa et al., 1996, 1997;

Perou et al., 1996b; Nagle et al., 1996). A helical region of Lyst was similar to stathmin,

a phosphoprotein that regulates the polymerization of microtubules and acts as a relay for

intracellular signal transduction. Therefore, this region of Lyst potentially encodes a

protein interaction domain that may regulate microtubule-mediated lysosome trafficking.

Although the CHS protein (LYST) is not predicted to have transmembrane helices, the

carboxy-terminal tetrapeptide (CYSP) of the mouse Lyst-II isoform is similar to known

prenylation sites and may provide attachment to membranes through thioester linkage

with the cysteine. A novel conserved domain, designated BEACH (BEige And CHs), was

defined based on homology with open reading frames from S. cerevisiae, C. elegans and









a human cell division control protein 4-related protein (CDC4L). The function of the

BEACH domain remains unknown. Helical regions in the CHS protein resemble ARM

and HEAT repeat motifs, which tend to form long rods. Some HEAT-containing proteins

are associated with vesicular trafficking but the functional significance of these motifs in

Lyst is unclear. The C-terminal region of the CHS protein contains several consecutive

WD40 motifs, which were originally identified in the P-subunit of the G-protein

transducin. This WD40 region may form P-sheets arranged in a propeller-like structure

that is thought to mediate protein-protein interactions. Based on the presence of ARM,

HEAT, BEACH and WD40 domains, it has been proposed that, similarly to the yeast

serine/threonine protein kinase VPS15, Lyst functions as a component of a membrane-

associated signal transduction complex that regulates intracellular protein trafficking. The

CHS protein contains multiple sites of potential phosphorylation by casein kinase II

(CKII), PKC, cAMP-dependent protein kinase and tyrosine kinase. Lyst seems to contain

helical bundles with clusters of phosphorylation sites at their ends. Phosphorylation of

these positions could provide a control mechanism by causing conformational change in

the bundles, thereby affecting interactions with other molecules. In summary, the

domains identified in Lyst support its potential involvement in the regulation of signal

transduction and protein trafficking but do not place the CHS protein in any known

family and do not define any proteins that interact with Lyst, leaving the question about

its mechanism of action unanswered.

Therefore, the main objective of this dissertation was to characterize the Chediak-

Higashi syndrome gene and its products in more detail, and to identify proteins that









interact with Lyst, in order to get some insights about its biological role. Following the

cloning of the CHS gene in human and mouse, the following specific aims for the present

dissertation were determined:

1. Identification of mutations in patients with Chediak-Higashi syndrome and in

beige mice.

2. Identification and characterization of Lyst mRNA isoforms.

3. Identification and characterization of Lyst2, a brain-specific member of the

Chediak-Higashi syndrome gene family.

4. Identification of proteins that interact with the CHS protein and with LYST2,

using a yeast two-hybrid approach.

Detection of mutations in the CHS gene would help the identification of critical

amino acids and domains in Lyst and would enable genotype-phenotype correlation, as

well as the design of tests for early diagnosis of Chediak-Higashi syndrome. The

characterization of different mRNA isoforms would determine the functional importance

of individual alternatively spliced variants and their expression in different tissues. The

identification of proteins that interact with Lyst would reveal the pathways in which the

CHS protein is involved and would enhance our knowledge about the mechanisms of

regulation of protein trafficking, particularly to and from the lysosomes.














MATERIALS AND METHODS


Reverse Transcription-Polymerase Chain Reaction (RT-PCR) Amplification of CHS

Patient and Beige Mouse cDNAs


Lymphoblastoid and fibroblast cell lines from CHS patients were obtained from

the Coriell Institute for Medical Research (Camden, NJ). Mice carrying different alleles

of the beige mutation and control mice were obtained from The Jackson Laboratory (Bar

Harbor, ME), sacrificed, and their organs were harvested and homogenized. Total cellular

RNA was isolated from human and mouse cells by extraction with phenol/guanidine

isothiocyanate (TRIzol Reagent, Life Technologies Inc., Gaithersburg, MD). One to three

p.g of RNA were reverse transcribed using SuperScript II RNase H- reverse transcriptase

(Life Technologies Inc.), with oligo(dT)-primed first strand cDNA synthesis. DNA

fragments of interest were PCR amplified with KlenTaq DNA polymerase (Clontech

Laboratories, Palo Alto, CA) using the following amplification conditions: 94C for 2

min, then 30 cycles of 94C for 20 s, 55C for 15 s, 68C for 1 min, followed by 6 min at

68C. PCR products were separated by electrophoresis on agarose gels, visualized by

ethidium bromide staining and purified using Wizard PCR Preps DNA Purification

System (Promega, Madison, WI). One tL of purified PCR product was used as template

for a hemi-nested PCR amplification with KlenTaq DNA polymerase under the same









conditions as above. PCR products were separated by electrophoresis on agarose gels,

visualized by ethidium bromide staining and purified using Wizard PCR Preps DNA

Purification System. The DNA concentration in each sample was determined using a

Spectronic 1001 spectrophotometer (Milton Roy/Bausch & Lomb, Rochester, NY) and

the purified PCR products were sequenced either directly or following TA cloning and

plasmid purification.




TA Cloning of PCR Products and Plasmid Purification


PCR products were cloned in the pCR 2.1 vector using the TA Cloning kit

(Invitrogen, San Diego, CA) according to the manufacturer's instructions. Briefly, PCR

products were ligated in pCR 2.1 vector overnight at 14C and INVaF' competent E. coli

cells were transformed with the constructs. Cells were grown overnight at 37C on LB

plates containing 50 plg/mL kanamycin and 40 mg/mL X-Gal, then liquid LB/kanamycin

cultures were inoculated with positive colonies and incubated overnight with shaking

(RPM=225) at 37C, and plasmids were isolated using the Wizard Plus Minipreps DNA

Purification System (Promega). Purified plasmids were visualized on ethidium bromide-

stained agarose gels, the DNA concentration in each sample was determined using a

Spectronic 1001 spectrophotometer and -0.3 [tg were used for sequencing.








Single-Strand Conformation Polymorphism (SSCP) Analysis


Detection of nucleotide changes by SSCP was performed as described by

Aberanthy et al., 1997. Briefly, each PCR product was mixed with an equal volume of

denaturing buffer and heated to 95C for 3 min. The samples were loaded onto 0.8 mm

thick, 10% native polyacrylamide gels which were run at ambient temperature at 9 W for

6-10 h, depending on the size of the PCR product. Bands were visualized by silver

staining (Beidler et al., 1982).




DNA Sequencing and Sequence Analysis


An ABI PRISM 310 Genetic Analyzer and dye terminator cycle sequencing kit

(Perkin Elmer, Foster City, CA) with AmpliTaq DNA polymerase, FS were used for

sequencing according to the manufacturer's instructions. Both strands of purified PCR

products and plasmids were sequenced to confirm the accuracy of the results. Sequences

were analyzed with the GCG Package (Devereux et al., 1984) and searches of the

National Center for Biotechnology Information database were performed using the

BLAST network server (Altschul et al., 1990) (National Library of Medicine, via

INTERNET) and the Whitehead Institute Sequence Analysis Program (MIT, Cambridge,

MA). For analysis of sequences, identified by the yeast two-hybrid method as potential

interacting proteins, see the "Yeast Two-Hybrid Screens" section later in this chapter.








Allele-Specific Oligonucleotide Analysis


PCR products spanning the mutation site in patient 371 were transferred to nylon

membranes using a slot-blot apparatus. Approximately 5 ng of each PCR product was

treated with a denaturing solution (0.5 M NaOH, 1.5 M NaCl), split in half and loaded in

duplicate. Two 17mer oligonucleotides that span the region containing the mutation were

synthesized. One contained the sequence of the normal allele (5'-CGCACATG

GCAACCCTT-3'), while the other contained the sequence of the mutant allele (5'-

GCACATGGGCAACCCTT-3'). These were end-labeled with [y-32P]dATP using T4

polynucleotide kinase and hybridized to the membranes at 50C. Hybridization and wash

buffers were as described (Church and Gilbert, 1984). Membranes were washed

sequentially at 45 C, 55 C and 65C for 10 min each and exposed to X-ray film.




PCR Amplification of Mouse Lyst cDNA Isoforms


Total cellular RNA was isolated from mouse bone marrow cells by extraction

with phenol/guanidine isothiocyanate (TRIzol Reagent). Five jIg of RNA were reverse

transcribed using SuperScript II RNase H reverse transcriptase, with first strand cDNA

synthesis priming by an oligo(dT) primer and a gene-specific primer (5'-CTCGCAGAG

CGGTGCTTATGTCCTGTG-3', 5'-CACAGTCATGGGACTGCTAA-3', for Lyst-I and

Lyst-II, respectively). One iL of cDNA was used as template for PCR amplification

using KlenTaq DNA polymerase. PCR conditions were: 94C for 1 min, then 30 cycles of

94C for 15 s, 68 C for 10 or 5 min (for Lyst-I and Lyst-II, respectively), followed by 20








or 10 min (for Lyst-I and Lyst-II, respectively) at 68 C. The same upstream primer (5'-

AGCGGAGGTGAAGCCTTATGCTGAGACAGT-3') was used for PCR amplification

of both isoforms. Isoform-specific downstream primers used were 5'-

TTATGTCCTGTGGGGACACTCCTTC-3' (for Lyst-I) and 5'-ACAGAGCATCC

CCACTTCCCTATCTAAAGT-3' (for Lyst-II). The 2 shorter mouse isoforms (Lyst-III

and Lyst-IV) were PCR amplified from mouse melanocyte cDNA using KlenTaq DNA

polymerase. PCR conditions were: 94C for 2 min, then 36 cycles of 94C for 20 s, 56C

for 15 s, 68C for 2.5 min, followed by 10 min at 68C. PCR products were separated by

electrophoresis on agarose gels and visualized by ethidium bromide staining. One p.L of

1:500 dilution of PCR products was used as template for a nested PCR amplification with

KlenTaq DNA polymerase. PCR conditions were: 94C for 2 min, then 33 cycles of 94C

for 20 s, 56C for 6 s, 68C for 3 min, followed by 6 min at 68C. PCR products were

separated by electrophoresis on 1.2% low melting point agarose gels, visualized by

ethidium bromide staining, excised from the gel with a sterile blade and purified using

Wizard PCR Preps DNA Purification System. Purified PCR products were cloned in pCR

2.1 vector by the TA method, and sequenced.




Cloning of Human LYST cDNAs


Normal human bone marrow poly(A) RNA was obtained from Clontech

Laboratories and 1 p.g was reverse transcribed using SuperScript II RNase H reverse

transcriptase, with a LYST-specific primer (5'-CTTGTTGGCTAGTGCATATTGA

CACAATCTTCC-3') used for first strand cDNA synthesis. The resulting LYST-specific








cDNA was used as a template for PCR amplification of the LYST coding domain with

KlenTaq DNA polymerase in 50 ptL reactions. PCR conditions were: 94C for 2 min, then

33 cycles of 94 C for 20 s, 63 C for 15 s (for some reactions) and 68C for 10 min,

followed by 12 min at 68C. PCR products were separated on ethidium bromide-stained

low melting point 1% agarose gels, excised with a sterile blade, purified using Wizard

PCR Preps Purification System, and cloned in pCR 2.1 vector by the TA method. Some

PCR products were treated with 10 U exonuclease I (United States Biochemicals) for 15

min at 37 C and reamplified by a nested PCR with KlenTaq DNA polymerase prior to

cloning in the pCR 2.1 vector. The identity of the cloned products was confirmed by PCR

amplification from purified plasmids with LYST-specific primers using Taq DNA

polymerase in 20-uL reactions. PCR conditions were: 94C for 2 min, then 30 cycles of

94C for 20 s, 56 C for 10 s, 72C for 1 min, followed by 5 min at 72 C. PCR products

were visualized on ethidium bromide stained agarose gels and photographed.

Segments of the human LYST cDNA sequence were obtained by an anchored,

nested PCR (5'RACE-PCR) using liver cDNA (Clontech Laboratories) as a template, by

RT-PCR using total RNA and by sequencing of human expressed sequence tags (EST)

similar in sequence to mouse Lyst. For the 5'RACE-PCR, 2 nested primers were derived

from a human EST (GenBank accession No. W26957) and had the following nucleotide

sequences: 5'-CCAAGATGAAAGCAGCCGATGGGGAAAACT-3' and 5'-

TCAGCCTCTTTCTT GCTCCGTGAAACTGCT-3'. For RT-PCR experiments, total

RNA was prepared from the promyelocytic HL-60 cell line. RT-PCR was performed with

Expand polymerase (Boehringer Mannheim) with the following primer pairs: 5'-








AGTTTATGAGTCCAAATGAT-3' and 5'-GAATGATGAAGTTGCTCTGA-3' (bp

490-2034); 5'-CAGCAGTTCTTCAGATGGA-3' and 5'-ATCTTTCTGTTGTTCCC

CTA-3' (bp 1891-3050); and 5'-TAGGGGAGCAACAGAAAGAT-3' and 5'-GCTCAT

AGTAGTATCACTTT-3' (bp 3320-4722). The primers used to amplify the cDNA

between base pairs 1891 and 3050 were derived from the mouse Lyst sequence. Human

primers were designed from the sequence of the PCR product (1159 bp) and used to

amplify the flanking sequences. Human LYST intron oa' was PCR amplified from human

genomic DNA (100 ng) with the primers 5'-CCGCTCGAGTAGGATCTT

TAAGGTGAATAAC-3' and 5'-GTGATACTACTATGAGCCCTTCACAGTATC-3'.

PCR conditions were: 94C for 1 min, then 32 cycles of 94 C for 15 s, 63 C for 20 s and

72 C for 30 s, followed by 10 min at 72 C.




Molecular Probes


The LYST2 probe was PCR amplified from a human fetal brain partial length

cDNA clone (#32273), obtained from Image Consortium, using internal primers.

KlenTaq DNA polymerase was used for amplification and the PCR conditions were as

follows: 94C for 2 min, then 36 cycles of 94 C for 30 s, 54 C for 15 s, 68 C for 5 min,

followed by 5 min at 68 C. PCR product was then purified with Wizard PCR Preps DNA

Purification kit. Other probes are described in the text or in the legends of the

corresponding figures. All probes were radio-labeled by the hexanucleotide technique

(Prime-It II random primer kit, Stratagene) with [at-32P]dCTP (Amersham) according to

the manufacturer's instructions.









Isolation ofLYST2 cDNA Clones


LYST2 clones were isolated by screening cDNA libraries using vector-insert PCR

and hybridization. A human fetal brain cDNA library (Life Technologies Inc.) was

screened using hemi-nested PCR with one vector primer and one LYST2-specific primer.

One p.L of cDNA library (5x109 cfu/ml) was used as template for PCR amplification with

KlenTaq DNA polymerase. PCR conditions were: 94C for 5 min, then 33 cycles of 94C

for 30 s, 63 C for 30 s, 68 C for 3.5 min, followed by 10 min at 68 C. PCR products

were separated by electrophoresis on agarose gels and visualized by ethidium bromide

staining. PCR products were excised from low melting agarose gels, purified (Wizard

PCR Preps DNA Purification kit), and cloned using the TA method. Resulting plasmids

were purified with Wizard Plus Minipreps DNA Purification kit and sequenced.

Membranes containing clones from a human fetal brain cDNA library (Clontech

Laboratories) were screened by hybridization with a radioactively labeled cDNA probe

from the 3'-end of the LYST2 coding domain. Positive clones were detected by

phosphorimaging (Molecular Dynamics), and purified plasmids were isolated and

sequenced. Then new probes were PCR amplified from the 5'-end of the novel cDNA

sequence and the library screening was repeated in order to isolate overlapping clones

extending the sequence in the 5' direction.

A ZAP Express EcoRI mouse embryo cDNA library was screened by

hybridization with the LYST2 probe. Identified cDNA clones were purified using Wizard

Plus Minipreps DNA Purification kit and sequenced.








Southern Blot Analysis and Autoradiography


Genomic DNA was isolated from mouse organs using standard techniques

(Sambrook et al., 1989) or obtained from The Jackson Laboratory (Bar Harbor, ME).

DNA was digested with restriction endonucleases from Promega (Madison, WI) or

Boehringer Mannheim (Indianapolis, IN), and 10-jig samples were subjected to

electrophoresis in 0.85% agarose gels for 20 h at 35V. DNA was transferred to

ZetaProbe (Bio-Rad Laboratories, Hercules, CA) or GenescreenPlus (DuPont Co.,

Wilmington, DE) membranes, which were neutralized in 2x SSC, UV-crosslinked for 1

min and prehybridized at 65C overnight. Hybridization to probes labeled by the

hexanucleotide technique (Prime-It II random primer kit, Stratagene, La Jolla, CA) with

[a-32P]dCTP (Amersham, Arlington Heights, IL) was performed at 65C overnight.

Membranes were washed in a solution containing 2x SSC, 0.1% SDS and 1 mM EDTA

for 30 min at 65C and then in a solution containing 0.5x SSC, 0.1% SDS and 1 mM

EDTA for 5-15 min at 65C. Blots were exposed to X-ray film (DuPont) in FisherBiotech

(Pittsburgh, PA) autoradiography cassettes for 1-7 days and films were developed using a

Konica QX60A processor.




Northern Blot Analysis


Isolation of poly(A) RNA from fibroblast and Epstein-Barr virus-transformed B

lymphoblastoid cell lines, formaldehyde agarose gel electrophoresis and northern blotting

to ZetaProbe membranes were performed according to standard procedures (Sambrook et








al., 1989). Northern blots of 2 [tg poly(A) RNA from various mouse tissues, human

cancer cell lines and human lymphoid tissues, as well as a dot-blot containing poly(A)

RNA from 50 human adult and fetal tissues were obtained from Clontech Laboratories.

The quantity of poly(A) RNA on each dot was normalized for 8 housekeeping genes,

and varied from 100 to 500 ng. Membranes were hybridized with various probes labeled

with [a-32p]dCTP by the hexanucleotide technique (Prime-It II random primer kit).

Hybridization conditions were the same as those employed with Southern blots.

Autoradiography or phosphorimaging was performed using X-ray film or

phosphorimager (Molecular Dynamics, Sunnyvale, CA), respectively.




Simple Sequence Length Polymorphism (SSLP) PCR Amplification


PCR reactions with Taq DNA polymerase (Boehringer Mannheim) were

performed using 40 ng of mouse genomic DNA, 1 jimol/L of each primer (Research

Genetics, Inc., Huntsville, AL) and 200 jimol/L of each dNTP in a 20 gL reaction. PCR

conditions were: 95C for 2 min, then 30-36 cycles of 94C for 20 s, 55 C for 30 s, 72 C

for 20 s, followed by 72 C for 1 min. Amplification products were separated on 3%

agarose gels, visualized by ethidium bromide staining and photographed.




Backcross Mouse Panel and Genetic Mapping


C57BL/6J-bg' X (C57BL/6J-bgo' x CAST/EiJ)F, backcross mice were bred and

maintained as described in Holcombe et al., 1991. Genomic DNA was isolated from









mouse organs using standard techniques (Sambrook et al., 1989) and Southern blots were

prepared. The chromosomal assignment of each probe was determined by linkage

analysis in 93 intersubspecific backcross mice. The 192 genetic markers (156

microsatellite markers, 14 genes, 20 expressed sequence tags (Tchemev et al., 1997) and

2 coat color loci) that have been previously mapped in this panel served as anchor loci

with average genetic distance between them 10 cM. Gene order was determined by

minimization of crossover events and elimination of double crossovers between linked

genes. Intergene distances were expressed in centimorgans (cM) and standard errors were

determined as previously described (Green, 1981; Bishop, 1985).




Cloning of LYST and LYST2 Bait cDNA Fragments in Yeast Two-Hybrid Vectors


Fourteen overlapping cDNA fragments (with average size of -850 bp each),

covering the entire coding domain of LYST, were PCR amplified from LYST cDNAs

cloned in the pCR 2.1 vector. LYST2 cDNA fragments were amplified from EST clone

#32273, obtained from Image Consortium. Primer design ensured that all cDNA

fragments would be amplified in frame and that known domains with potential functional

significance would be preserved. All primers contained specific restriction sites at their

5'-ends and all reverse primers contained stop codons near their 5'-ends. The sequence of

all forward and reverse primers with their restriction sites, and the length and the position

(according to GenBank accession number U67615 for LYST) of the amplified cDNA

fragments are shown in Table 2 (for LYST) and in Table 3 (for LYST2).






Table 2. PCR primers, with restriction sites on their 5'-ends, used to amplify bait cDNA fragments from the LYST gene

GENE, PRIMER SEQUENCE, 5'-> 3' RESTR. FRAGMENT FRAGM
FRAGMENT # (forward upper sequence; reverse lower sequence) ENZYME POSITION,bp SIZE,bp
LYST GCGGTGGATCCCATGAGCACCGACAGTAACTCAC BamHI 190-1056 866
1 GCGGAATTCTCATAGTGTGGGCACTACACTGG EcoRI 9-6_
LYST GCGGTGGATCCCACACTAACTGAGTTCCTAGCAGGCTTTGGGGACTGC BamHI 1051-1950 899
2 (GAGAGA)3CTCGAGTCAATCCATACAACAGCATATTCCAATG XhoI
L YST GCGGAACTAGTGGATCCCAAATCTGTAATCATTCC SpeI 1948-2370 422
3 GGAATTCTCACTGAACAACTATATTGCCTTTCTG EcoRI
L YST GCGAGTGTGGATCCCCAGAAAGGCAATATAGTTGTTCAG BamH 2I
4 GCGGAATTCTCACTCCTCTTTGTGACTTCTGAAC EcoRI
LYST GCGGTGGATCCCCTGTTCAGAAGTCACAAAGAGG BamHI 3190-4032 842
5 GGAATTCTCAAGCAGAAAGCAAATTTAATTCCAG EcoRI
LYST GCGGTGGATCCCCTGGAATTAAATTTGCTTTCTGC BamHI 4009-4821 812
6 GCGGAATTCTCAGGAGCCCAGTGAAATTATATG EcoRI
LYST GCGGTGGATCCCTCCAAAGCGTTGATGATCCAAG BamHI
7 GCGGAATTCTCAAACTCGCAGTGCTAATGCTTG EcoRI 419-5
LYST GCGGTGGATCCCACTCAAGCATTAGCACTGCGAG BamHI 5677-6612 935
8 (GAGAGA)3CTCGAGTCAGGCAACATAAGTATCTGCAATATTTTG XhoI
LYST GCGGTGGATCCCCAAAATATTGCAGATACTTATGTTGCC BamHI 6586-7449 863
9 GCGGAATTCTCATCCCATGTTTCTCACATCTTCCAG EcoRI
LYST GCGGTGGATCCCCTGGAAGATGTGAGAAACATGGG BamHI 7426-8238 812
10 GCGGAATTCTCAGGTCTGGAAAACTGAGGTCTTG EcoRI _
LYST GCGGTGGATCCCAAGACCTCAGTTTTCCAGACCG BamHI 8218-9039 821
11 GCGGAATTCTCAATCCAACTGCCATGAGGTTGG EcoRI
LYST GCGGTGGATCCCGATCCAACAGAAGGGCCAAATC BamHI 9037-9585 548
12 GCGGAATTCTCATGTCAGAGCGGTGATGTTACC EcoRI
LYST GCGGTGGATCCCAAGGTTCGTGATGATGTATACCAC BamHI 9502-10590 1088
13 GCGGAATTCTCACAAGCCTTTTATCCATGACAAAGG EcoRI I_21__
LYST GCGGTGGATCCCTGGATAAAAGGCTTGAAATGGGG BamHI 10576-11611 1035
14 GCGGAATTCTCATGAAGTTCATTCGCATTCACCC EcoRI I_ __ _









Table 3. PCR primers, with restriction sites on their 5'-ends, used to amplify bait cDNA fragments from the LYST2 gene ___
GENE, PRIMER SEQUENCE, 5'-+ 3' RESTR. FRAGMENT FRAGM.
FRAGMENT # (forward upper sequence; reverse lower sequence) ENZYME POSITION, bp* SIZE, bp
L YST2 GCGGAACTAGTGACTTCTGATGTAAAGGAAC SpeI (3-794) 791
1 GCGGAATTCTCAATAGCGATTATCTGCGTGTAC EcoRI (374) 791
LYST2 GCGGAACTAGTGACTTCTGATGTAAAGGAAC SpeI (3-686) 683
2 GCGGAATTCTCATGGATCCATTTCAATGGG EcoRI (3-66)
LYST2 GCGGAACTAGTGGTAACAGCAGATAATCGCTAT SpeI (774-1424) 650
3 GCGGAATTCTCACCAGGGTTAAATGTACAGTTG EcoRi (4424)
LYST2 GCGGAACTAGTGACTTCTGATGTAAAGGAAC Spel 141
4 GCGGAATTCTCACCAGGGTTAAATGTACAGTTG EcoRI (3-1424) 1421


* Since the entire LYST2 cDNA sequence is not known, fragment positions are relative to the beginning of the known LYST2
sequence.









KlenTaq DNA polymerase and 12.5 pmol of each primer were used in each 50 JIL

amplification reaction. PCR conditions were: 94C for 5 min, then 30 cycles of 94 C for

40 s, 50 C for 40 s, 72 C for 2 min, followed by 72 C for 5 min. PCR products were

visualized on ethidium bromide-stained agarose gels, purified using QIAquick PCR

purification kit (QIAGEN Inc., Valencia, CA) and digested with the corresponding

restriction enzymes (Tables 2 and 3) in 20 iiL-reactions overnight at 37 C. Digested PCR

products were excised from 1% low melting point agarose gels, purified using QIAquick

gel extraction kit (QIAGEN Inc.) and ligated in the yeast Gal4 activation domain cloning

vector pAD-GAL4 (modified pGAD GH, Clontech Laboratories) and in the yeast Gal4

DNA-binding domain vector pGB-GAL4 (modified pGBT9, Clontech Laboratories).

DH 10OB competent cells were transformed with the purified constructs by electroporation,

incubated for 1 h at 37 C with shaking (RPM=150) and plated on LB/ampicillin (50

p.g/mL) plates containing X-Gal and IPTG. The identity of cloned fragments was

confirmed by PCR of bacterial colonies with one vector primer and one fragment-specific

primer under the same amplification conditions as above. LB/ampicillin liquid cultures

were inoculated with positive colonies and incubated overnight at 37 C with shaking.

Purified plasmids were isolated using QIAprep Spin Plasmid kit (QIAGEN Inc.) and

visualized on ethidium bromide-stained agarose gels. All inserts were sequenced to

ensure that PCR amplification reproduced an accurate copy of the LYST and LYST2

sequence and that the cDNA fragments were cloned in frame.









Yeast Two-Hybrid Screens


Two types of screening for protein-protein interactions were performed by the

yeast two-hybrid method. In a "forward" screen, one hybrid consisted of the DNA

binding domain of the yeast transcriptional activator Gal4 fused to a "bait" portion of

LYST or LYST2. The other hybrid consisted of the Gal4 activation domain fused to

"prey" sequences encoded by a mammalian cDNA library. In a "reverse" screen, the

"bait" part of LYST or LYST2 was fused to the Gal4 activation domain, and the "prey"

sequences of the mammalian cDNA library were fused to the Gal4 DNA binding domain.

The prey cDNAs were obtained from a human fetal brain cDNA library of 1 x 107

independent isolates (Clontech Laboratories). The library was synthesized from

XhoI-(dT)15-primed fetal brain mRNA (pooled from five male/female 19-22 week fetuses)

that was directionally cloned into either pAD-GAL4, an yeast Gal4 activation domain

cloning vector including the LEU2 gene for selection in yeast deficient in leucine

biosynthesis, or pBD-GAL4, an yeast Gal4 DNA-binding domain cloning vector

including the TRP1 gene for selection in yeast deficient in tryptophane biosynthesis.

To ensure that the bait DNA-binding domain fusions do not possess intrinsic

transcriptional activity, a test for "bait self-activation" was performed. YULH yeast cells,

transformed by the lithium acetate/polyethylene glycol method (Ito et al., 1983) with the

pBD-GAL4 bait constructs, were grown at 30C on SC-Trp medium to select for the

presence of the DNA-binding domain plasmid. The selected yeast cells were then plated

on SC-Ura medium, grown overnight at 30C, and examined for growth. Cells that did

not grow contained baits which were not "self-activating" proteins, that is, these proteins









required interaction with a second protein domain in order to form a functional complex

that can activate the transcription of reporter genes. Baits from non-growing cells were

used for further screening.

In the forward screens, the constructs encoding the baits were introduced by

lithium acetate/polyethylene glycol transformation into the yeast strain YULH (mating

type a, ura3, his3, lys2, Ade2, trpl, leu2, gal4, gal80, GAL1-URA3, GAL]-lacZ), while

the prey sequences were transformed into the yeast strain N106r (mating type a, ura3,

his3, ade2, trpI, leu2, gal4, gal80, cyhr, Lys2::GALlUAS-HIS37A7A-HIS3,

ura3::GALlJu GALTATA-lacZ). For the reverse screens, baits were transformed into N106r

and preys into YULH. Initially, a limited screening using approximately 50,000 prey

library members was performed to ensure that bait fusions do not interact promiscuously

with many proteins (test for non-specific binding). Only baits that did not possess

intrinsic transcriptional activity (i.e., passed the test for "bait self-activation") and that did

not interact indiscriminately with numerous proteins (i.e., passed the test for non-specific

binding), were used for screening the entire prey library. In these full-scale screens, the

two transformed populations were mated using standard methods (Sherman et al., 1991).

Briefly, cells were grown at 30C until mid-to-late log phase on media that selected for

the presence of the appropriate plasmids, i.e. yeast containing activation domain

constructs were grown on SC-Leu medium, while cells containing DNA binding domain

plasmids were grown on SC-Trp medium. The two mating strains, a and a, were then

diluted in YPAD media (Sherman et al., 1991), filtered onto nitrocellulose membranes,

and incubated at 30C for 6-8 hours. The cells were then transferred to media selective for









the desired diploids, i.e., for yeast harboring reporter genes for beta-galactosidase, uracil

auxotrophy, and histidine auxotrophy, and for expression of the vectors encoding the bait

and prey. The mating products were plated on SC (synthetic complete) media (Kaiser et

al., 1994) lacking adenine and lysine (to select for successful mating), leucine and

tryptophane (to select for expression of genes encoded by the bait and prey plasmids),

and uracil and histidine (to select for protein interactions). This medium is referred to as

SCS medium, for SC Selective medium.

Selected clones were tested for expression of P3-galactosidase to confirm the

formation of an LYST:LYST-interacting protein (IP), resp. LYST2:LYST2-IP,

interactions. Filter-lift p3-galactosidase assays were performed as modified from the

protocol of Breeden and Nasmyth, 1985. Colonies were patched onto SCS plates, grown

overnight, and replicated onto Whatman No. 1 filters. The filters were then assayed for

P3-galactosidase activity. Colonies that were positive turned a visible blue, indicating that

a protein-protein interaction has occurred, enabling the transcription of the reporter gene

LacZ.

Cells in colonies positive for protein interaction contained both DNA-binding and

activation-domain plasmids. These cells were regrown as single isolates in individual

wells of 96-well plates. Ten microliters of each isolate were lysed, the inserts within the

pAD-GAL4 and pBD-GAL4 plasmids were amplified by PCR using primers specific for

the insert-flanking sequences of the vectors, and approximately 300 amino-terminal bases

of each insert were determined using an ABI 377 sequencer. Comparison to known

sequences was made using the BLAST program publicly available through the National









Center for Biotechnology Information (N.C.B.I.). The assembly and identity searches of

the sequences encoding human EST were performed by using publicly available EST

assembly databases such as the N.C.B.I. BlastN 2.0 program (Altschul et al, 1990).

Sequences that aligned with 95% or greater identity at the nucleic acid level over their

termini of at least 30 nucleotides were utilized if the alignment resulted in 5' extension or

3' extension of the EST sequence. Once this first assembly was complete, the extended

sequence was again subject to the BlastN comparison to detect new homologies to the

added extensions. The sequence was extended in both directions until new related

sequences that allowed extension of the assembled sequence were no longer detected. The

assembled EST sequence was subjected to further searches using the BlastX 2.0 program

for the identification of protein coding regions by database similarity search (Gish and

States, 1993). The BlastX software translates the DNA sequence in all six reading frames

and compares the translated protein sequence with those in protein databases. The

statistical significance is estimated under the assumption that the equivalent of one entire

reading frame in the query sequence codes for protein and that significant alignments will

involve only coding reading frames. Furthermore, the sequences were analyzed for open

reading frames using software that translates the assembled DNA sequence in all six

reading frames using the standard genetic code. The interacting ESTs were obtained from

directionally-cloned libraries, and thus the direction of translation of the assembled EST

is known as 5' to 3' and the open reading frame is also known. ORFs longer than 50

amino acids following an initiator codon or an ORF with no initiator methionine encoded

at the 5'-end were determined to be possible protein products, and were compared to









sequences in protein data bases using the BLASTP 2.0 program (Altschul et al., 1990).

Further protein sequence analysis was performed after selecting a suitable open reading

frame. The protein sequence was compared to previously characterized protein domains

present in the BLOCKS and PRODOM motif databases (Bairoch, 1992; Henikoff and

Henikoff, 1991; Nakai and Kanehisa, 1992; Wallace and Henikoff, 1992).




Confirmation of the Specificity of Protein Interactions by the Yeast Two-Hybrid System


In addition to sequencing all baits, and the tests for "bait self-activation" and for

non-specific binding, described in the previous section, several other tests were

performed to determine the specificity of the detected bait:prey interactions. In the test for

"prey self-activation", YULH yeast cells were transformed with individual plasmids

encoding the DNA-binding domain fusions of all preys that were detected to interact with

LYST or LYST2. The transformed haploid yeast cells were grown at 30C on SC-Trp

medium to select for the presence of the DNA-binding domain plasmid. The selected

yeast cells were plated on SC-Ura medium, grown overnight at 30C, and examined for

growth. Cells that did not grow contained preys which were not "self-activating" proteins,

that is, these proteins required interaction with a second protein domain in order to form a

functional complex that can activate the transcription of reporter genes. Preys from non-

growing cells were used in the other confirmation tests.

YULH and N106r haploid yeast cells, transformed with DNA-binding domain,

resp. activation domain fusions of preys that were detected to interact with LYST or

LYST2, were grown on SC-Trp or SC-Leu media to select for the corresponding








plasmids. Filter-lift P3-galactosidase assay was performed on the selected haploid cells.

Colonies that did not turn blue contained preys that needed to interact with another

protein in order to activate the transcription of the LacZ reporter gene. Only cells that did

not grow on SC-Ura medium (i.e., passed the test for "prey self-activation") and that did

not turn blue (i.e., passed the P-galactosidase test for haploids) were used in the next

confirmation tests.

The following "matrix mating" test was performed to duplicate the initial forward

and reverse screens and to determine whether the same protein-protein interactions would

be detected. YULH yeast cells, transformed with DNA-binding domain fusions of baits

(forward screen) or preys (reverse screen), and N106r cells, transformed with activation

domain fusions of baits (reverse screen) or preys (forward screen), were grown in YPAD

medium at 30C. The "matrix mating" was performed by spreading yeast containing

individual baits on YPAD plates, transferring (in 96-well format) prey-containing cells

onto the spread baits, and incubating the plates overnight at 30C. Each plate contained

also positive (proteins that are known to interact) and negative (proteins that do not

interact with a particular bait) controls. Resulting diploid colonies were replica-plated on

2 types of media: on SCS, which selected for successful mating, for protein-protein

interactions and for presence of plasmids, and on SC-Leu-Trp plates, which were used for

a filter-lift 13-galactosidase assay. Only baits and preys from diploid yeast cells that were

able to grow on SCS medium (i.e., passed the "matrix mating" test) and that turned blue

(i.e., passed the P3-galactosidase test for diploids), were PCR amplified with vector

primers and the resulting PCR products were sequenced. If sequencing determined that











the identity of a particular bait:prey pair matched that obtained after the initial screens,


the interaction was considered confirmed by the yeast two-hybrid method. An overview


of the described yeast two-hybrid screening and confirmation procedures is shown in


Figure 2.




CLONE LYSTAND LYST2 BAITS IN
ACTIVATION- AND BINDING-DOMAIN VECTORS
4,
SEQUENCE ALL CONSTRUCTS
4,
TRANSFORM IN YEAST
4' 4,
TEST FOR TEST FOR
BAIT SELF-ACTIVATION NON-SPECIFIC BINDING
4, 4,
FORWARD AND REVERSE SCREENS
4,
P3-GAL. TEST FOR DIPLOIDS
4,
PCRINSERIS
4'
SEQUENCE INSERTS
4,
DETERMINE IDENTITY OF BAIT:PREY INTERACTIONS
4,
RE-TRANSFORM YEAST
4, 4,
TEST FOR PREY SELF-ACTIVATION P3-GAL. TEST FOR HAPLOIDS
4, 4,
MATRIX MATING
4,
P3-GAL. TEST FOR DIPLOIDS
4,
PCR INSERTS
4,
SEQUENCE INSERTS
4,
DETERMINE IDENTITY OF BAIT:PREY INTERACTIONS
4,
COMPARE TO INITIAL IDENTITIES
4,
INTERACTION CONFIRMED BY THE YEAST TWO-HYBRID METHOD

Figure 2. Overview of the screening and confirmation procedures by the yeast
two-hybrid method.














RESULTS


Identification of Mutations in Patients with Chediak-Higashi Syndrome and in Beige

Mice


Detection of mutations in the CHS gene of human patients and beige mice was

undertaken for three main reasons:

To provide additional evidence that LYST, resp. Lyst is indeed the gene

responsible for the Chediak-Higashi syndrome in human and for the beige

mutation in mouse

To identify critical amino acids and domains that are essential to the function

of the CHS gene

To determine whether a genotype-phenotype correlation could be established

between different genetic defects and specific disease manifestations.

A summary of mutations detected in patients with Chediak-Higashi syndrome is

shown in Table 4. Several different approaches were used for mutation detection. As an

initial screen for mutations, we analyzed northern blots of poly(A) RNA from CHS

patients. The largest mRNA species (-13 kb) was greatly reduced in abundance or absent

in lymphoblastoid mRNA of patients P 1 and P3, respectively (Fig. 3A), while the smaller

transcript (-4.4 kb) was present and undiminished in abundance in all three patients.










B


4N 4


12kb-

4.4kb-


C


imA> It


373


Figure 3. Mutation analysis in CHS patients. (A) Northern blot of 2 jg aliquots of
lymphoblastoid poly(A) RNA from CHS patients and a control. The probe used for
hybridization corresponds to nucleotides 490-817 of the CHS cDNA (GenBank accession
No. U70064). Exposure time for autoradiography was 48 h. (B) SSCP analysis of cDNA
corresponding to nucleotides 439-806 of the CHS cDNA. Each lane contains samples
from individual patients as indicated. Extra bands in lanes corresponding to patients 371
and 373 are denoted with an arrow. (C) Sequence chromatograms showing mutations in
cDNA clones from patients 371 and 373. The upper part is normal human CHS cDNA
sequence. The arrows indicate the positions of a G insertion (coding domain nucleotide
118, patient 371), and C-+T substitution (coding domain nucleotide 148, patient 373).
The sense strand is shown.








While the selective loss of the larger transcript in these two patients cannot be

excluded definitively, rehybridization of this blot with an actin probe suggested that the

absence of the larger transcript was not due to uneven gel loading or RNA degradation.

Fibroblast poly(A) RNA from three other CHS patients (369, 370 and 373) showed a

moderate reduction in the -13 kb mRNA (51-60% of control by densitometry), while the

-4.4 kb mRNA was essentially unaltered in abundance (103-147% of control).

Single-strand conformation polymorphism (SSCP) analysis was undertaken using

cDNA samples derived from lymphoblastoid or fibroblast cell lines from CHS patients, in

collaboration with Drs. Steve Colman and Margaret Wallace. Anomalous bands were

detected in PCR products from the 5'-end of the open reading frame in two unrelated

CHS patients (371 and 373) that were different from those with probable selective loss of

the larger transcript on northern blots (Fig. 3B). Sequence analysis identified a C--T

substitution at nucleotide 148 of the coding domain in patient 373 (Fig.3C). Four of the

nine cDNA clones derived from patient 373 contained this mutation. Restriction enzyme

digestion confirmed the mutation: TaqI digestion of cDNA (nucleotides 520-808) showed

loss of this restriction site in patient 373 to be heterozygous. The C-+T substitution

creates a stop codon at amino acid 50 (R50X). The mutation in patient 373 occurs within

a CpG dinucleotide, which, when the cytosine is methylated, represents a common

hotspot for mammalian mutations due to deamination.

In patient 371, a G insertion was found at nucleotide 118 of the coding domain,

resulting in a frame shift at codon 40 and termination after amino acid 62 (Fig. 3C).

While analysis of genomic DNA had shown patient 371 to be heterozygous for this









mutation (Nagle et al., 1996), each of five cDNA clones isolated from lymphoblasts of

this patient contained the G insertion. Allele-specific oligonucleotide hybridization of

cDNA from this patient failed to detect a signal with an oligonucleotide corresponding to

the normal allele, suggesting that the patient is transcriptionally hemizygous for the G

insertion. Alternatively, the lack of allele-specific oligonucleotide hybridization may

reflect a base mismatch (such as a polymorphism) within the patient's cDNA.

Mutations were identified in three other CHS patients by cloning and sequencing

fragments of their LYST coding domains. A homozygous C->T substitution at nucleotide

3085 of the coding domain creating a stop codon at amino acid 1029 (Q1029X) was

found in patient 370. In patient 369, a heterozygous frameshift mutation was detected.

Nucleotides 3073 and 3074 of the coding domain were deleted in two of five cDNA

clones, resulting in a frameshift at codon 1026 and termination at codon 1030. A

homozygous C->T substitution at nucleotide 3310 of the coding domain was identified in

patient 372. This mutation created a stop codon at amino acid 1103, resulting in

premature termination. Despite mutation analysis of >6 kb of open reading frame, the

mutations in CHS patients P1 and P3, that may have a selective loss of the larger

transcript on northern blots, have not been identified.

Lymphoblasts from all of these patients (369, 370, 371, 373, P1 and P3) contain

the giant perinuclear lysosomal vesicles that are the hallmark of the Chediak-Higashi

syndrome. The parents of patients 369 and 370 are known not to have been

consanguineous. Patients 369, 370 and 371 had typical clinical presentations of CHS,

with recurrent childhood infections and oculocutaneous albinism. In contrast, the clinical








course of patient 373 was milder this patient has not had systemic infections and

remains alive at age 37. Patient 373 does, however, have hypopigmented hair and irides,

as well as peripheral neuropathy. A summary of the described mutations, detected in

patients with Chediak-Higashi syndrome, is shown in Table 4.

Table 4. Mutations identified in patients with Chediak-Higashi syndrome.


PATIENT ZYGOSITY MUTATION CONSEQUENCE

,,. - p .,. ,, Codon 1026 frameshift,
369 Heterozygous bp 3073 & 3074 deleted Codon 1026 frameshift,
stop at codon 1030



370 Homozygous C3085T Q1029X


Codon 40 frameshift,
371 Heterozygous G insertion at bp 118 Codon 40 frameshift,
stop at codon 62



372 Homozygous C3310T stop at codon 1103



373 Heterozygous C148T R50X


X .^,,,Codon 489 frameshift,
Nagle et al., 1996 Homozygous bp 1467 deleted Codon 489 frameshift,
Nagle et aL., 1996 '*stop at codon 566



PI ? 13.5 kb mRNA lack of protein



P3 ? 13.5 kb mRNA lack of protein











Mutations were also identified in several beige alleles. A 5-kb genomic deletion

that contained the 3'-end ofLyst exon P3, and exons y and 6, was found in bg"' DNA. This

mutation was detected by hybridization of Lyst-derived probes to Southern blots of

genomic DNA from all potential progenitor mouse strains (Fig. 4), and by PCR

amplification of several Lyst fragments (Fig. 5). This deletion corresponds to a loss of

approximately 400 internal amino acids of the predicted Lyst protein. Furthermore,

whereas the 5'-end of the bg"' deletion occurs within Lyst exon 3, the 3'-end is intronic.

Therefore the truncated Lyst mRNA in bg"j mice is also anticipated to splice incorrectly,

terminate prematurely, and lack polyadenylation. Unfortunately, it was not possible to

confirm this prediction since bg"j mice are extinct and only archived genomic DNA was

available.

Quantitative reverse transcription (RT)-PCR demonstrated a moderate reduction

in Lyst mRNA in bg and bg' liver, and a gross reduction in bge (Lyst AOD after

normalization for P3-actin mRNA: +/+, 1.00; bgj/bgj, 0.19; bg/bg, 0.28; bg'/bg', 0.40)

(Fig. 6). A commensurate reduction in bg2e transcript abundance was noted by using

several primer pairs derived from different regions of the Lyst cDNA. Aberrant Lyst RT-

PCR products were not observed. The particularly striking (more than fivefold) reduction

in Lyst expression evident in bgj homozygotes suggested the existence of a mutation in

bgZ that results in decreased transcription or mRNA instability. The molecular basis of

the reduction in Lyst mRNA in bge is not yet known.













A B C


9.4-
6.6-~
4.36- -



2.2-
2.0- -m

1.4- -
1.1- -
0.9- "" <
0.6- -

HindIII Mspl Taql Hind III Mspl TaqI Hind Ill Msp Taql









Figure 4. Southern blot identification of an intragenic Lyst deletion in bg"j. A
Southern blot was sequentially hybridized with 3 Lyst probes: (A) the probe (nucleotides
1262-3433 of Lyst cDNA) extends upstream of the bg"j deletion. (B) the probe
(nucleotides 2835-3433 of Lyst eDNA) is completely deleted; (C) the probe (nucleotides
3594-4237 of Lyst cDNA) extends downstream of the bg"' deletion. Restriction
endonucleases are indicated at the bottom of each panel, and molecular size standards (in
kb) are shown to the left. Southern blots were prepared from genomic DNA of all
potential progenitor mouse strains, but only C57BL/10J, C57BL/6J and C57BL/6J- bg'jJ
are shown.









A+

'+ '0e


1.4-
1.1-
0.9-

0.6 -


B

CID


C A


s -
t's- P


S1.9-kb


1.6-kb 0.9-kb 2


bg11J deletion--


.1-kb


-i 3
03


Figure 5. PCR analysis of the bg"' deletion and genomic structure of Lyst in the
vicinity of the deletion. C57BL/10J, C3HeB/FeJ, C57BL/6J and C57BL/6J- bg"j
genomic DNA and Lyst cDNA were used templates in the PCR reactions. Amplicons
illustrated correspond to: (A) Lyst cDNA nucleotides 1337-1837, which represent exon 3
and are upstream of the deletion; (B) nucleotides 2670-3210, which represent exon y,
deleted in bg"j DNA; (C) nucleotides 4913-5433, which represent an exon downstream
from the deletion. No amplicon was observed in control PCR reactions performed
without template. More than 30 other STSs that had been localized within the bg non-
recombinant interval amplified normally from bg"' DNA. (D) Genomic structure of Lyst
in the vicinity of the bg"' deletion. Lyst exons (ac, 1, y, 6, c and <) are depicted by black
boxes, and intervening introns by a solid line. Nucleotides of the mouse Lyst cDNA that
correspond to exonic boundaries are indicated above the boxes. The 3' end of exon P3, and
all of exons y and 6, are deleted in bg"j DNA. Genomic structure and intronic sequences
were ascertained by sequence analysis of nested PCR products, performed with exonic
primers and P1 genomic clone as template. Boundaries of the bg"j deletion were
determined by PCR of genomic DNA.


*B


19















Lyst


Actb

% of +/+
(normalized)


100 40


19 28


Figure 6. Quantitative reverse transcription (RT)-PCR analysis of Lyst in several
bg alleles. Lyst mRNA from bg', bgj, bg and control livers was reverse transcribed, PCR
amplified and the amount of the resulting PCR products was quantitated by measuring
AOD after normalization for P-actin mRNA (% AOD for Lyst are shown below the gel
image).


X\


^ ^ 0











Identification and Characterization of Lyst mRNA Isoforms


The investigation of mRNA isoforms of the CHS gene was undertaken for the

following reasons:

To resolve the discrepancy between the initial reports about the identification

of the beige gene. The partial cDNA sequence, published by Perou et al.,

1996b, did not match the complete open reading frame, reported by Barbosa et

al., 1996.

To identify the isoform of primary functional significance, as well as other

alternatively spliced variants that may have some importance.

To determine which of the predicted domains are crucial for the function of

the CHS protein.

The existence of more than one mRNA isoform of the mouse beige gene

(designated Lyst) had been suggested by northern analysis (Fig. 7) and by the two

different Lyst cDNA sequences that were reported. Northern blots indicated that Lyst is

ubiquitously transcribed, both temporally and spatially, in mouse and human tissues.

Complex alternative splicing, with both constitutive and anatomically restricted Lyst

mRNA isoforms, was also revealed. The largest Lyst transcript in human and mouse was

12-13 kb and additional smaller alternatively spliced variants were also present with

varying abundance in different tissues (Fig. 7).









A


9.5-
7.5-
4.4-

2.5-
1.4-











9.5-
7.5-
4.4-

2.5-
1.4-


B

AAs4y *


3' Lyst Mid. Lyst

C D


x o


5' LYST


5' LYST


Figure 7. Northern blot analysis of mouse and human LYST. (A), (B): northern
blots of 2 jtg poly(A) mRNA from various mouse tissues hybridized with probes that
correspond to (A), nucleotides 4423-4631, and (B), nucleotides 1430-2457 (exon 3) of
mouse Lyst cDNA. (C), (D): northern blots of 2 jLg poly(A) mRNA from (C), human
lymphoid tissues, and (D), human cancer cell lines, hybridized with a probe that
corresponds to nucleotides 357-800 of human LYST cDNA. Molecular size standards (in
kb) are shown to the left.








Mouse Lyst cDNA isoforms were identified by anchored nested PCR (3'-RACE-

PCR). Two fragments (1.25 kb and 2 kb) were amplified from mouse spleen cDNA using

this technique. The 1.25 kb clone contained the 3'-end of the previously described 5893

bp Lyst cDNA that corresponds to a small mRNA isoform (Lyst-II). The 5'-end of the 2

kb clone, however, contained sequence derived from the Lyst-II cDNA, while the 3'-end

sequence was from the largest isoform (Lyst-I, previously called BG) of the beige gene.

Reverse transcription (RT)-PCR confirmed that nucleotides 1-4706 of Lyst are common

to both mRNA isoforms (Fig. 8A). The large isoform cDNA that contains the entire

coding domain was assembled from nucleotides 1-4706 of Lyst, the 2 kb 3'RACE-PCR

clone, and 6824 nucleotides of BG cDNA. This 11 817 bp cDNA sequence (Lyst-I,

GenBank accession number U70015) corresponds to the largest mRNA observed on

northern blots (Fig. 7). This -11.8 kb cDNA however, is truncated at both the 5'-end of

the 5'-untranslated region (UTR) and at the 3'-end of the 3'-UTR, and thus is smaller

than the largest mRNA (-12-13 kb), observed on mouse northern blots.

Analysis of a P1 genomic clone (number 8592), containing the entire beige gene,

revealed that the 11,817 bp Lyst-I cDNA results when splicing occurs from Lyst exon a

(containing nucleotide 4706) to downstream exon T (Figs. 8 and 10). In contrast,

incomplete splicing and reading through intron c' (interposed between exons a and r)

yields the truncated Lyst-II isoform, of length 5893 bp cDNA (Lyst-II, Fig. 8, GenBank

accession number L77884). Lyst intron c' encodes 37 in-frame amino acids followed by a

stop codon and a polyadenylation signal. Lyst-II corresponds to one of several smaller

mRNAs observed on northern blots (Fig. 7). Full-length cDNAs corresponding to the








Lyst-l and Lyst-II isoforms were both amplified from mouse bone marrow RNA by RT-

PCR. The putative Lyst-I and Lyst-II proteins are of relative molecular mass 425 287 (MK

425 kDa) and Mr 172.5 kDa, respectively.


a 477 b

Lyst-I 11.8kb IN
1078
Exon-I Exon-II 72
Intron-I'


Lyst-II Lyst-I
RNA nNA RNA -r


Lyst-l 5.9kb


F1/R1 FI/R2


Figure 8. Alternative splicing of the CHS gene. (A) Alternative splicing of mouse
Lyst. Solid lines represent Lyst exons a and T (not drawn to scale). Splicing of exon aC to
exon -r occurs in the Lyst-I mRNA (-12 kb). The hatched box represents the intronic
region that forms the 3' end of the Lyst-II ORF (5.9 kb). The intron contains a stop codon
(*) and a polyadenylation signal (A). Nucleotide positions indicated are from GenBank
accession No. L77884 (Lyst-II) and U70015 (Lyst-I). (B) Detection of mRNA isoforms of
mouse Lyst by RT-PCR and genomic PCR. DNase-treated mouse melanocyte RNA was
reverse transcribed and amplified with primers F 1/R1 (expected amplicon size 273 bp) or
Fl/R2 (expected amplicon size 560 bp). RNase-treated C57BL/6J DNA was amplified
with primers F 1/R 1.


47.06









cDNAs corresponding to the human homolog of the largest mRNA isoform of the

beige gene were obtained by identification of human expressed sequence tags (ESTs)

similar in sequence to mouse Lyst-I by database searches (GenBank accession numbers

L77889, W26957 and H51623). Intervening cDNA sequences were isolated using RT-

PCR, and a partial human cDNA sequence (GenBank accession number U70064; 7.1 kb)

was assembled by alignment of these clones with the largest mouse bg gene cDNA. The

predicted human and mouse peptides shared 82% identity over 1990 amino acids. The

predicted human amino acid sequence contains a six amino acid insertion at residue 1039,

relative to that of mouse. The complete sequence of the human CHS cDNA was reported

by Nagle et al., 1996. This 13.5 kb cDNA sequence corresponds to the largest mRNA

observed on northern blots from human tissues (CHS-I, Fig. 7 and inset of Fig. 9). These

northern blots also demonstrated the existence of a smaller human transcript (4.5 kb) that

was similar in size to the small mouse Lyst mRNA, and that appeared to differ from the

large isoform in distribution of expression in human tissues (Fig. 7). Assuming that the

genomic derivation of the small human isoform might be the same as the small mouse

isoform, the 3'-end of the small human isoform was identified by cloning human intron

or' using PCR of human genomic DNA with primers derived from exon a and intron ac'.

Similarly to intron a' in the mouse, the sequence of the 5'-end of human intron c'

contained 17 codons in frame with exon a, followed by a stop codon (CHS-II, inset of

Fig. 9). cDNA corresponding to this short human isoform was amplified from human

peripheral blood RNA by RT-PCR with primers from a 5' exon and from intron a',

indicating that this intron was indeed present in CHS-II mRNA.















IU

9 CHS-I -13kb WD40
5' 3'

Stathmin-like .. Probe 2


t 6 Probe 1 Intron o'
Uo
Z
w 5 CHS-II 6kb
z4.

3-








n0 0
W ,O







Figure 9. Comparison of the relative abundance of CHS (LYST) mRNA isoforms
in 50 human tissues after normalization. Probe 2 (specific for the largest isoform CHS-I,
or LYST-I) and the corresponding signal intensity levels are shown in black. Probe 1
(identifies all CHS transcripts) and the corresponding signal intensity levels are shown in
white. Variation of the relative mRNA abundance among tissues was assessed by
phosphorimaging, dot quantitation, subtraction of background and calculation of the
intensity percentage for each tissue. The relative mRNA abundance between normalized
tissues for each probe was expressed as the percentage of the total hybridization signal
obtained for each tissue with that probe. The average percentage intensity for
housekeeping genes in all tissues was -2%. No hybridization was evident to control dots,
containing yeast total RNA, yeast tRNA, E. coli rRNA, E. coli DNA, poly r(A), human
Cotl DNA and human DNA. Exposure time for phosphorimaging was 20 h.









Nucleotides 1-5905 of human small and long isoform cDNAs are identical, and

are followed either by intron a' sequence in the short isoform (GenBank accession

number U84744), or by exon T and the rest of the exons, in the long isoform (inset of Fig.

9). The predicted intron-encoded amino termini of the short mouse and human isoform

peptides share 65% identity.

Additional splicing complexity of smaller isoforms exists. The Lyst-Ill isoform

lacks exons a and 13, while in Lyst-IV, exons a, 13 and y are absent. In both isoforms the

alternative splicing occurs out of frame and results in termination shortly after the splice

junction, suggesting that most likely these two isoforms do not have functional

significance. These splice variants correspond to additional bands observed on northern

blots (Fig. 7). A schematic representation of the four Lyst isoforms is shown in Fig. 10.





exons
Lyst-1,~-12kb 5'1 a 1 cP3 157 1Y1E 1 3'

Lyst-I,~-5.9kb JP 17 1 l I al

Lyst-IlI, 720 bp P INTRONIC

Lyst-IV, 744 bp Ml1
*



Figure 10. Schematic representation of the four mRNA isoforms of the Chediak-
Higashi syndrome gene (*, stop codon; A, poly-A).








In an effort to evaluate the functional significance of the two largest Lyst

isoforms, genetic complementation experiments were initiated in collaboration with Dr.

Stephen Brandt, Vanderbilt University. Full-length Lyst-I and Lyst-II isoforms were

amplified from mouse bone marrow cDNA by long-range PCR and cloned in expression

vectors. The resulting constructs were expressed in cultured fibroblasts from beige mice,

exhibiting the characteristic giant perinuclear granules in the cytoplasm. The effect was

evaluated by observing the changes in the size and distribution of specifically stained

lysosomes and endosomes. Some cells, expressing full-length Lyst-I, demonstrated more

peripheral distribution of the stained granules in the cytoplasm and slight reduction of

their size. Such changes were not visible in cells expressing the smaller Lyst-II isoform,

supporting the hypothesis that the largest splice variant (Lyst-I) represents the mRNA

isoform of primary functional significance. However, since the effects of the

complementation were not very prominent and were not observed in all cells expressing

Lyst-I, the significance of the described genetic complementation is questionable.

Analysis of northern blots of mouse mRNA had suggested that the relative

abundance of the large and small transcripts differed from tissue to tissue (Fig. 7). The

relative abundance of mRNA isoforms of the homologous gene in human tissues at

different developmental stages was examined (Fig. 9) by sequential hybridization of a

poly(A) RNA dot-blot with a cDNA probe (inset of Fig. 9, Probe 2, shown in black,

nucleotides 10941-11590 of human CHS, GenBank accession No U67615) specific for

the largest isoform and with a cDNA probe (inset of Fig. 9, Probe 1, shown in white,

nucleotides 190-445 of human CHS, GenBank accession No U70064 and U84744) that

identifies all CHS transcripts (Fig. 9). The inset in Fig. 9 shows the human 5.8 kb cDNA









isoform (CHS-II, or LYST-II, which arises through incomplete splicing, with intron o'

supplying the 3'-end of the transcript), the largest isoform (CHS-I, or LYST-I, 13.5 kb,

which results from removal of intron a') and the location of the probes. The quantity of

poly(A) RNA on the blot was normalized to eight housekeeping genes (phospholipase,

ribosomal protein S9, tubulin, highly basic 23 kDa protein, glyceraldehyde-3-phosphate

dehydrogenase, hypoxantine guanine phosphoribosil transferase, P-actin and ubiquitin).

Using a probe that hybridized only to the largest mRNA isoform on northern

blots, transcripts were most abundant in thymus (adult and fetal), peripheral blood

leukocytes, bone marrow and several regions of the adult brain (Fig. 9). Interestingly, the

largest mRNA isoform was not detected in fetal brain. There was also low relative

expression of this isoform in heart, lung, kidney and liver in both adult and fetal tissues.

A somewhat different pattern of relative expression was evident upon

rehybridization of the blot with a probe derived from the 5'-end of the coding domain of

the CHS gene, a region that hybridized to all mRNA isoforms on northern blots.

Consistent with the transcription pattern of the largest isoform, this probe detected

abundant expression on peripheral blood leukocytes, thymus (adult and fetal) and bone

marrow. However, several tissues with abundant large isoform transcripts exhibited

considerably less relative expression with the 5'-probe, including most regions of the

adult brain, fetal and adult thymus, and spleen. Furthermore, several tissues with

negligible relative transcription of the large isoform exhibited increased relative

expression with the 5'-probe, including adult and fetal heart, kidney, liver, and lung; adult

aorta, thyroid gland, salivary gland and appendix; and fetal brain (Fig. 9).








Identification and Characterization of Lyst2. a Brain-Specific Member of the Chediak-

Higashi Syndrome Gene Family


Identification and Sequence Analysis

In an effort to identify genes similar to the CHS gene, extensive database searches

were performed. BLASTX comparison of mouse Lyst (U70015) to public EST databases

revealed a significant, but non-identical, match with an uncharacterized human EST

(R17955). The corresponding partial-length cDNA clone (#32273, length 1986 bp),

derived from a human infant brain cDNA library, was sequenced. An additional 250 bp of

cDNA sequence, located immediately 5' of clone 32273, was obtained by insert-vector

PCR of a human fetal brain cDNA library. The resulting total sequence (2.2 kb, Fig. 11 a)

corresponded to the 3' end of the coding domain and the 3' UTR of a novel gene that was

designated LYST2 (lysosomal trafficking regulator 2, Fig.1 lb). Amino acid identity of

LYST2 with LYST varied from region to region: LYST2 amino acids 16-222 shared 50.2

% identity with LYST residues 3215-3425. This region corresponds to the BEACH

domain (amino acids 3116-3461) of LYST, and exhibited sequence similarity to

anonymous ORFs from S. cerevisiae, C. elegans and human CDC4L protein (Nagle et

al., 1996). In contrast, the WD repeat-containing carboxy-terminal domain of LYST,

which is predicted to assume a six-bladed beta-propeller structure similar to the B-subunit

of heterotrimeric G proteins (Sondek et al., 1996), shared little amino acid similarity with

LYST2. The latter region of LYST2 did, however, contain LYST-like WD repeats, and

also exhibited significant sequence similarity to the B-subunit of heterotrimeric G proteins








(30.4 % identity LYST2 amino acids 368-501 to GB {P49027}). The LYST 3' UTR

demonstrated 42.1% identity to 584 bp of the 3' UTR of LYST2.

In an effort to identify the mouse homolog of human LYST2, a mouse embryo

cDNA library was screened with a human LYST2 probe, and two overlapping clones of

total length of-2.5 kb were identified (Fig. l ic). The corresponding partial mouse

putative protein (Lyst2, Fig. 1 ld) exhibited 29.4% identity over 676 amino acids to

mouse Lyst and 96.2 % identity in 557 amino acids to human LYST2.

Genetic Mapping

In order to confirm that Lyst and Lyst2 are indeed two different genes, and to

evaluate the potential candidacy of Lyst2 for mouse coat color mutations, Lyst2 was

mapped by cross-hybridization in the mouse using unique CAST/EiJ restriction fragment

length polymorphisms with MspI (-4.1 kb) and EcoRI (-3.6 kb). Linkage analysis using

DNA from 93 intersubspecific backcross [C57BL/6J-bg X (C57BL/6J-bg' x CAST/EiJ)

F1] mice revealed that Lyst2 maps on mouse Chr 3 (Fig. 12). The best gene order and

recombination frequency ( standard error) were: centromere D3MUt21 6.452.5 cM -

Lyst2 5.382.3 cM D3Mit22-telomere (GenBank accession number AF072372).

LYST2 was mapped on human Chr 13 by hybridization to Southern blots of

human-rodent somatic cell hybrids, in collaboration with Dr. Margaret Wallace. No

cross-hybridization to LYST was observed (Fig. 13). While the CHS gene maps on human

Chr 1 and on mouse chromosome 13, LYST2 was localized to human Chr 13 and to

mouse chromosome 3, confirming that LYST and LYST2 are indeed two different genes.







63



a) Human LYST2 DNA:
1 CCGAAGAGAG CTGTGTTTTA TGCAGAGCGT TATGAGACAT GGGAAGATGA
51 TCAAAGCCCA CCCTACCATT ATAATACCCA TTATTCAACA GCAACATATA
101 CTTTATCCTG GCTTGTTCGA ATTGAACCTT TCACAACCTT CTTCCTCAAT
151 GCAAATGATG GAAAATTTGA TCATCCAGAT CGAACCTTCT CATCCGTTGC
201 AAGGTCTTGG AGAACTAGTC AGAGGGATAC TTCTGATGTA AAGGAACTAA
251 TTCCAGAGTT CTACTACCTA CCAGAGATGT TTGTCAACAG TAATGGATAT
301 AATCTTGGAG TCAGAGAAGA TGAAGTAGTG GTAAATGATG TTGATCTTCC
351 CCCTTGGGCA AAAAAACCTG AAGACTTTGT GCGGATCAAC AGGATGGCCC
401 TAGAAAGTGA ATTTGTTTCT TGCCAACTTC ATCAGTGGAT CGACCTTATA
451 TTTGGCTATA AGCAGCGAGG ACCAGAAGCA GTTCGTGCTC TGAATGTTTT
501 TCACTACTTG ACTTATGAAG GCTCTGTGAA CCTGGATAGT ATCACTGATC
551 CTGTGCTCAG GGAGGCCATG GAGGCACAGA TACAGAACTT TGGACAGACG
601 CCATCTCAGT TGCTTATTGA GCCACATCCG CCTCGGAACT CTGCCATGCA
651 CCTGTGTTTC CTTCCACAGA GTCCGCTCAT GTTTAAAGAT CAGATGCAAC
701 AGGATGTGAT AATGGTGCTG AAGTTTCCTT CAAATTCTCC AGTAACCCAT
751 GTGGCAGCCA ACACTCTGCC CCACTTGACC ATCCCCGCAG TGGTGACAGT
801 GACTTGCAGC CGACTCTTTG CAGTGAATAG ATGGCACAAC ACAGTAGGCC
851 TCAGAGGAGC TCCAGGATAC TCCTTGGATC AAGCCCACCA TCTTCCCATT
901 GAAATGGATC CATTAATAGC CAATAATTCA GGTGTAAACA AACGGCAGAT
951 CACAGACCTC GTTGACCAGA GTATACAAAT CAATGCACAT TGTTTTGTGG
1001 TAACAGCAGA TAATCGCTAT ATTCTTATCT GTGGATTCTG GGATAAGAGC
1051 TTCAGAGTTT ATACTACAGA AACAGGGAAA TTGACTCAGA TTGTATTTGG
1101 CCATTGGGAT GTGGTCACTT GCTTGGCCAG GTCCGAGTCA TACATTGGTG
1151 GGGACTGCTA CATCGTGTCC GGATCTCGAG ATGCCACCCT GCTGCTCTGG
1201 TACTGGAGTG GGCGGCACCA TATCATAGGA GACAACCCTA ACAGCAGTGA
1251 CTATCCGGCA CCAAGAGCCG TCCTCACAGG CCATGACCAT GAAGTTGTCT
1301 GTGTTTCTGT CTGTGCAGAA CTTGGGCTTG TTATCAGTGG TGCTAAAGAG
1351 GGCCCTTGCC TTGTCCACAC CATCACTGGA GATTTGCTGA GAGCCCTTGA
1401 AGGACCAGAA AACTGCTTAT TCCCACGCTT GATATCTGTC TCCAGCGAAG
1451 GCCACTGTAT CATATACTAT GAACGAGGGC GATTCAGTAA TITCAGCATF
1501 AATGGGAAAC TITI'GGCTCA AATGGAGATC AATGATTCAA CACGGGCCAT
1551 TCTCCTGAGC AGTGACGGCC AGAACCTGGT CACCGGAGGG GACAATGGGG
1601 TAGTAGAGGT CTGGCAGGCC TGTGACTTCA AGCAACTGTA CATTTAACCC
1651 TGGATGTGAT GCTGGCATTA GAGCAATGGA CTTGTCCCAT GACCAGAGGA
1701 CTCTGATCAC TGGCATGGCT TCTGGTAGCA TTGTAGCTTT TAATATAGAT
1751 TTTAATCGGT GGCATTATGA GCATCAGAAC AGATACTGAA GATAAAGGAA
1801 GAACCAAAAG CCAAGTTAAA GCTGAGGGCA CAAGTGCTGCATGGAAAGGC
1851 AATATCTCTG GTGGAAAAAA TTCGTCTACA TCGACCTCCG TTTGTACATT
1901 CCATCACACC CAGCAATAGC TGTACATTGT AGTCAGCAAC CATTTTACTT
1951 TGTGTGTTTT TTCACGACTG AACACCAGCT GCTATCAAGC AAGCTTATAT
2001 CATGTAAATT ATATGAATTA GGAGATGTTT TGGTAATTAT TTCATATATT
2051 GTTGTITATT GAGAAAAGGT TGTAGGATGT GTCACAAGAG ACTTTTGACA
2101 ATTCTGAGGA ACCT'GTGTC CAGTTGTTAC AAAGTTTAAG CTTTGAACCT
2151 AACCTGCATC CCATIVCCAG CCTCTTTTCA AGCTGAGAAA AAAAAAAAAA
2201 AAAAA

b) Human LYST2 peptide:
1 PKRAVFYAER YETWEDDQSP PYHYNTHYST ATYTLSWLVR IEPFTTFFLN
51 ANDGKFDHPD RTFSSVARSW RTSQRDTSDV KELIPEFYYL PEMFVNSNGY
101 NLGVREDEWVV VNDVDLPPWA KKPEDFVRIN RMALESEFVS CQLHQWIDLI
151 FGYKQRGPEA VRALNVFHYL TYEGSVNLDS ITDPVLREAM EAQIQNFGQT
201 PSQLLIEPHP PRNSAMHLCF LPQSPLMFKD QMQQDVIMVL KFPSNSPVTH
251 VAANTLPHLT IPAVVTVTCS RLFAVNRWHN TVGLRGAPGY SLDQAHHLPI
301 EMDPLIANNS GVNKRQITDL VDQSIQINAH CFVVTADNRY ILICGFWDKS
351 FRVYTTETGK LTQIVFGHWD VVTCLARSES YIGGDCYIVS GSRDATLLLW
401 YWSGRHHIIG DNPNSSDYPA PRAVLTGHDH EVVCVSVCAE LGLVISGAKE
451 GPCLVHTITG DLLRALEGPE NCLFPRLISV SSEGHCIIYY ERGRFSNFSI
501 NGKLLAQMEI NDSTRAILLS SDGQNLVTGG DNGVVEVWQA CDFKQLYI*

c) Mouse Lyst2 DNA:
1 GCAGCAGGGC GAACCGGACC TCTGTGATGT TTAATTTTCC TGACCAAGCA
51 ACAGTTAAAA AAGTTGTCTA CAGCTTGCCT CGGGTTGGAG TGGGGACCAG
101 CTATGGTTrG CCACAAGCCA GGAGGATATC ACTGGCCACT CCTCGACAGC
151 TGTATAAGTC TTCCAATATG ACTCAGCGCT GGCAAAGAAG GGAAATCTCC
201 AACTTrGAGT ATTrGATGTT TCTCAACACG ATAGCAGGTC GGACGTATAA
251 TGATCTGAAC CAGTATCCTG TGTTTCCATG GGTGTTAACA AACTATGAAT







64


301 CAGAGGAGTT GGACCTGACT CTCCCAGGAA ACTTCAGGCA TCTGTCAAAG
351 CCAAAAGGTG CTTTGAACCC GAAGAGAGCA GTGTTTTACG CAGAGCGCTA
401 TGAGACATGG GAGGAGGATC AAAGCCCACC CTTCCACTAC AACACACATT
451 ACTCAACGGC GACTTCCCCC CTTTCATGGC TTGTTCGGAT TGAGCCATTC
501 ACAACCTTCT TCCTCAATGC AAATGATGGG AAATTTGACC ATCCAGACCG
551 AACCTTCTCA TCCATTGCAA GGTCATGGAG AACCAGTCAG AGAGATACAT
601 CCGATGTCAA GGAACTAATT CCAGAGTTCT ATTACGTACC AGAGATGTTT
651 GTCAACAGCA ATGGGTACCA TCTTGGAGTG AGGGAGGACG AAGTGGTGGT
701 TAATGATGTG GACCTGCCCC CCTGGGCCAA GAAGCCAGAA GACTTTGTGC
751 GGATCAACAG GATGGCCCTG GAAAGTGAAT TTGTTTCTTG CCAACTCCAT
801 CAATGGATTG ACCTTATATT TGGCTACAAA CAGCGAGGGC CAGAGGCAGT
851 CCGTGCTCTC AATG'T1TCC ACTACTTGAC CTACGAAGGC TCTGTAAACC
901 TGGACAGCAT CACAGACCCT GTGCTCCGGG AGGCCATGGT TGCACAGATA
951 CAGAACTTTG CCCAGACGCC ATCTCAGTTG CTCATTGAGC CGCATCCGCC
1001 TAGGACTTCA GCCATGCATC TGTGTTCCCT TCCACAGAGC CCACTCATGT
1051 TCAAAGATCA GATGCAGCAG GATGTGATCA TGGTGCTGAA GTTTCCATCC
1101 AATTCTCCTG TGACTCATGT GGCTGCCAAC ACCCTGCCCC ACCTGACCAT
1151 CCCTGCAGTG GTGACAGTGA CCTGCAGCCG ACTGTTTGCA GTGAACAGAT
1201 GGCACAACAC AGTCGGCCTC AGAGGAGCCC CCGGATACTC CTTGGATCAA
1251 GCACACCATC TTCCCATTGA GATGGACCCA TTAATCGCAA ATAACTCTGG
1301 TGTGAACAAG CGGCAGATCA CAGACCTTGT AGACCAGAGC ATCCAGATCA
1351 ATGCCCACTG CTTCGTGGTC ACAGCTGATA ATCGCTACAT CCTCATCTGT
1401 GGGTTTTGGG ATAAAAGTTT CAGAGTTTAC TCGACAGAAA CAGGGAAACT
1451 GACACAGATT GTATTTGGCC ACTGGGATGT TGTCACATGC CTGGCCAGGT
1501 CGGAGTCCTA CATTGGTGGA GACTGCTACA TAGTGTCTGG ATCTCGGGAC
1551 GCCACCTTGC TTCTCTGGTA CTGGAGTGGG CGTCACCACA TCATCGGAGA
1601 CAACCCCAAT AGCAGTGACT ATCCTGCGCC CAGAGCTGTC CTCACAGGCC
1651 ATGACCATGA AGTTGTCTGT GTCTCCGTCT GTGCAGAACT CGGACTCGTT
1701 ATCAGTGGTG CTAAAGAGGG CCCTUGCCTC GTTCATACCA TCACTGGAAA
1751 TCTGCTGAAG GCCCTGGAAG GACCAGAAAA CTGCTTATTT CCACGCCTAA
1801 TTTCGGTATC CAGTGAAGGC CACTGCATCA TATATTATGA GCGAGGACGG
1851 TTTAGCAACT TCAGCATCAA TGGGAAACTT TTGGCTCAAA TGGAGATCAA
1901 TGATTCCACT AGGGCTATTC TCCTGAGCAG CGATGGACAG AACCTGGTGA
1951 CTGGAGGGGA CAATGGTGTG GTGGAGGTCT GGCAGGCCTG TGACTTTAAG
2001 CAGCTGTACA TTTACCCAGG ATGTGATGCT GGCATTAGAG CGATGGATTT
2051 ATCCCATGAC CAAAGGACTC TGATCACTGG CATGGCTTCC GGCAGCATTG
2101 TACTTTTAAT ATAGATTTTA ATCGGTGGCA TTATGAGCAT CAGAACAGTA
2151 CTGAAGAGAA GCAGCAGAAG CCACATTCAA GTGAGAGCAC AAGTGCTTCT
2201 GTGGAAAGGC AGTATCTCTG GTGGGACGCT GGTCCACATC GGCCTCTGCT
2251 TGTACATCCA TCCCACCCAG CAGTCGCCGA ACATCATAGT CGGGAGCCAT
2301 TTCACCCTGT TTTTCCAGGA CTGAACACCA GCTGCTGTCA AGCAAGCTTA
2351 TATCATGTAA ATTATCTGAA TTAGGAGCCG TTTTGGTAAT TATTTCATAT
2401 ATCGCCGTTT ATTGAGAAAA GGTTGTAGGA AGCCTCACAA GAGACTITTG
2451 ACAATTCTGA GGAACCTTGT GCCCAGTTGT TACAAAGTTT AAGCTTTGAA
CCTAACTTGC ATCCCATTTC CAGCCTCGGG CTTCACTCGT GCC

d) Mouse Lyst2 peptide:
1 SRANRTSVMF NFPDQATVKK VVYSLPRVGV GTSYGLPQAR RISLATPRQL
51 YKSSNMTQRW QRREISNFEY LMFLNTIAGR TYNDLNQYPV FPWVLTNYES
101 EELDLTLPGN FRHLSKPKGA LNPKRAVFYA ERYETWEEDQ SPPFHYNTHY
151 STATSPLSWL VRIEPFTTFF LNANDGKFDH PDRTFSSIAR SWRTSQRDTS
201 DVKELIPEFY YVPEMFVNSN GYHLGVREDE VVVNDVDLPP WAKKPEDFVR
251 INRMALESEF VSCQLHQWID LIFGYKQRGP EAVRALNVFH YLTYEGSVNL
301 DSITDPVLRE AMVAQIQNFA QTPSQLLIEP HPPRTSAMHL CSLPQSPLMF
351 KDQMQQDVIM VLKFPSNSPV THVAANTLPH LTIPAVVTVT CSRLFAVNRW
401 HNTVGLRGAP GYSLDQAHHL PIEMDPLIAN NSGVNKRQIT DLVDQSIQIN
451 AHCFVVTADN RYILICGFWD KSFRVYSTET GKLTQIVFGH WDVVTCLARS
501 ESYIGGDCYI VSGSRDATLL LWYWSGRHHI IGDNPNSSDY PAPRAVLTGH
551 DHEVVCVSVC AELGLVISGA KEGPCLVHTI TGNLLKALEG PENCLFPRLI
601 SVSSEGHCII YYERGRFSNF SINGKLLAQM EINDSTRAIL LSSDGQNLVT
651 GGDNGVVEVW QACDFKQLYI YPGCDAGIRA MDLSHDQRTL ITGMASGSIV
LLI*


Fig. 11. Partial cDNA (a, c) and predicted amino acid (b, d) sequences of human
(a, b) and mouse (c, d) lysosomal trafficking regulator 2, LYST2.


















D3Mit2 l [] DEWED
Lyst2 F] DDF EED
D3Mit221[ [I [ D

#of mice3745 4 2 4 1


D3Mit21


Lyst2


D3Mit22.


-cM 19.2
6.452.5 cM
5.382.3 cM
-cM33.7


EcoRP


Figure 12. Genetic mapping of Lyst2 on mouse chromosome 3.
a) Autoradiograph of a Southern blot of mouse genomic DNA, digested with
EcoRI and hybridized to a Lyst2 probe. Molecular size standards (in kilobases) are shown
to the left. The arrowhead indicates the RFLP, which was used for the genetic mapping
and was present in DNA from (C57BL/6J-bg' x CAST/EiJ) F, (be) but not in
homozygous C57BL/6J-bgj (bb) DNA.
b) Haplotype analysis of Lyst2, D3Mi21 and D3Mit22 on mouse chromosome 3
in 93 [C57BL/6J-bg X (C57BL/6J-bg x CAST/EiJ) FI] intersubspecific backcross mice.
Black squares represent the homozygous C57BL/6J pattern and white squares, the F1
pattern. The number of mice of each haplotype is indicated.
c) Composite genetic map of part of mouse chromosome 3, showing the relative
position of the Lyst2 locus. Marker locations are expressed as genetic distance from the
centromere (in centimorgans), and intermarker distances ( standard error) are shown.


bbbc
9.4 -*,g,
6 6 ^
4.4 T-

2.3













E -g E
-r 2 22 13 14 15 16 17 -
9.4--

6.6- I . .

4.3-











Figure 13. Mapping of LYST2 on human chromosome 13 by human-rodent
somatic cell hybrids. An EcoRI monochromosomal somatic cell hybrid Southern blot was
hybridized to [a-2P]dCTP-labeled LYST2 probe. Lane 1 contains control human DNA;
lane 2 control mouse DNA; lane 3 human Chr 22 on hamster background; lanes 4 to 8
- human Chr 13, 14, 15, 16, and 17, respectively, on mouse background; lane 9 control
hamster DNA. Molecular size standards (in kilobases) are shown to the left of the gel.








Expression Analysis

In order to determine the size of the LYST2 transcripts) and their abundance in

different organs, expression analysis was performed using northern blots from various

human and mouse tissues.

Transcript size in mouse and human. Hybridization of a LYST2 probe to northern

blots of mouse and human tissues (Fig. 14) revealed that the largest band was 12-13 kb,

very similar to the size of the largest isoform (LYST-I) of the Chediak-Higashi syndrome

gene (Barbosa et al., 1997). Low levels of additional transcripts of -6 kb and -5 kb were

visible in mouse brain RNA.

Transcript abundance in mouse. Hybridization to poly(A) northern blots from 8

mouse tissues showed that Lyst2 was abundantly expressed only in brain. Moderate

expression was observed in kidney, and weak expression in heart, lung, skeletal muscle

and testis. The gene was not expressed in spleen and liver (Fig. 14a).

Transcript abundance in human. In northern blots of selected human tumor cell

lines (Fig. 14b), LYST2 was moderately expressed in melanoma and colorectal

adenocarcinoma, weakly expressed in HeLa, lymphoblastic leukaemia and Burkitt's

lymphoma cells and not expressed in promyelocytic and myelogenous leukaemia, or in

lung carcinoma lines. In 7 normal human tissues, LYST2 was weakly expressed in spleen,

lymph node, thymus and appendix and not expressed in peripheral blood leukocyte, bone

marrow and fetal liver (Fig. 14c).

The relative abundance of LYST and LYST2 mRNA in 50 human tissues was

compared by sequential hybridization of corresponding gene-specific probes to a








poly(A) RNA dot blot (Fig. 15). The quantity of poly(A) RNA loaded on the blot was

normalized to 8 housekeeping genes (phospholipase, ribosomal protein S9, tubulin, a

highly basic 23 kDa protein, glyceraldehyde-3-phosphate dehydrogenase, hypoxanthine

guanine phosphoribosil transferase, 13-actin and ubiquitin).

When a probe specific for the largest mRNA isoform of LYST was hybridized to

the blot, transcripts were most abundant in immune system organs (adult and fetal

thymus, peripheral blood leukocytes, bone marrow) and in several regions of adult brain

(Fig. 15). There was moderate to low relative LYST expression in most of the other

tissues tested. Interestingly, no LYST expression was detected in fetal brain, as well as in

salivary gland, kidney, lung and fetal heart.

Following hybridization of the same blot with a LYST2-specific probe, a different

pattern of expression was observed (Fig. 15). LYST2 transcripts were most abundant in

whole brain (adult and fetal) and in 14 specific adult brain regions, as well as in kidney,

while there was low to negligible LYST2 expression in the rest of the tissues.


















1 2 34 5678


1234 5 6 7 8


C




1234567


'4


9.5-
7.5-

4.4-

2.5-

1.4-


Mouse


9.5-
7.5-

4.4-

2.5-

1.4-


Human


Human


Figure 14. Northern blot analysis of mouse and human LYST2.
Northern blots of 2 tg poly(A) RNA from various mouse tissues (a), human
cancer cell lines (b), or human lymphoid tissues (c), hybridized to a [a-32P]dCTP-
labeled LYST2 probe. Lanes: (a) 1-heart, 2-brain, 3-spleen, 4-lung, 5-liver, 6-
muscle, 7-kidney, 8-testis; (b) 1-promyelocytic leukemia, 2-HeLa, 3-
myelogenous leukemia, 4-lymphoblastic leukemia, 5-Burkitt's lymphoma, 6-
adenocarcinoma, 7-lung carcinoma, 8-melanoma; (c) 1-spleen, 2-lymph node, 3-
thymus, 4-appendix, 5-leukocyte, 6-bone marrow, 7-fetal liver. Molecular size
standards (in kb) are shown to the left.


9.5-
7.5-
4.4-

2.5-

1.4-





















S6
u)
Z
zU 5
uJ



3

2 i i






s I 8 i !










Figure 15. Comparison of the relative abundance of LYST and LYST2 transcripts
in 50 human tissues after normalization. Variation of the relative mRNA abundance
among tissues was assessed by phosphorimaging, dot quantitation, subtraction of
background and calculation of the intensity percentage for each tissue. The relative
mRNA abundance between normalized tissues for each probe was expressed as the
percentage of the total hybridization signal obtained for each tissue with that probe. The
average percentage intensity for housekeeping genes in all tissues was -2%. No
hybridization was evident to control dots, containing yeast total RNA, yeast tRNA, E.
coli rRNA, E. coli DNA, poly r(A), human Cotl DNA and human DNA. Exposure time
for phosphorimaging was 20 h.









Identification of Proteins that Interact with the CHS Protein and with LYST2. Using a

Yeast Two-Hybrid Approach


In an effort to investigate the function of the LYST and LYST2 proteins, two

types of screens for protein-protein interactions were performed by the yeast two-hybrid

method, as described in MATERIALS AND METHODS. In a "forward" screen, one

hybrid consisted of the DNA binding domain of the yeast transcriptional activator Gal4

fused to a "bait" portion of LYST or LYST2. The other hybrid consisted of the Gal4

activation domain fused to "prey" sequences encoded by a mammalian cDNA library. In

a "reverse" screen, the "bait" part of LYST or LYST2 was fused to the Gal4 activation

domain, and the "prey" sequences of the mammalian cDNA library were fused to the

Gal4 DNA binding domain. The activation domain vector contained the LEU2 gene for

selection in yeast deficient in leucine biosynthesis, and the DNA-binding domain cloning

vector included the TRP1 gene for selection in yeast deficient in tryptophane

biosynthesis. Each of the vectors was transformed into complementary (a and a) mating

types of yeast. Mating was carried out to express both vector constructs within the same

yeast cells, thus allowing interaction to occur. Interaction between the bait and prey

proteins led to transcriptional activation of reporter genes containing cis-binding elements

for Gal4. The reporter genes encoding the indicator protein beta-galactosidase, and

metabolic markers for uracil and histidine auxotrophy, were included in specific fashion

in one or the other of the yeast strains used in the mating, as described in the

MATERIALS AND METHODS chapter. As a result, yeast were selected for expression

of both fusion constructs, for successful mating, and for expression of LYST- or LYST2-








interacting proteins (IPs). Yeast clones that contained interacting proteins were grown in

individual wells of microtiter plates and the plasmids containing the IP-sequences were

isolated and characterized. Numerous tests (for bait self-activation, prey self-activation,

non-specific binding, sequencing of all baits, matrix mating, P3-galactosidase tests for

haploids and for diploids) were performed, as described in the MATERIALS AND

METHODS chapter, to determine the specificity of the detected bait:prey interactions.

The following interactions, summarized in Table 5, passed all confirmation tests.

In the forward screens, four different molecules were found to interact with the

CHS protein. LYST fragment 190-1056 interacted with sequences identical to the 14-3-3

protein. Three independent isolates of 14-3-3 sequence, starting at nucleotides 270, 306,

and 330, representing the same region of the molecule (as shown in APPENDIX,

Sequence 1), were identified. Another sequence which was found to interact with LYST

fragment 3190-4032 was identical to casein kinase II P13-subunit. Four independent isolates

of CKII-P3, starting at nucleotides 1, 4, 31, and 82 (APPENDIX, Sequence 3) and

representing the same region of the molecule, were identified. LYST fragment 6586-7449

was found to interact with a sequence identical to human hepatocyte growth factor-

regulated tyrosine kinase substrate (Hrs), starting at nucleotide 326 (APPENDIX,

Sequence 4). Another identified sequence interacting with LYST fragment 9037-9585

was identical to the EST cg51287.dlO sequence (IP-11, APPENDIX, Sequence 23),

starting at nucleotide 1.

In the reverse screens, a total of 22 molecules were identified as LYST- or

LYST2-IPs. Ten of those were identical to published proteins and thirteen molecules








were identified as EST sequences (eleven novel and two known ESTs). Identified

sequences interacting with the LYST fragment 190-1056 were identical to 14-3-3 protein

sequence starting at nucleotides 270, 288 and 354, i.e. representing the same domain of

14-3-3 protein identified in the forward screen (APPENDIX, Sequence 1), to B 14-3-3

protein sequence starting at nucleotide 466 (APPENDIX, Sequence 2), to calmodulin

sequence starting at nucleotide 474 (APPENDIX, Sequence 5), to estrogen receptor-

related protein, starting at nucleotide 756 (as shown in APPENDIX, Sequence 6), and to

EST cg50173.dlO sequence (IP-7, APPENDIX, Sequence 19), starting at nucleotide 1.

The sequence identical to 14-3-3 protein, starting at nucleotide 270, also interacted with

the LYST fragment 4009-4821. LYST fragment 6586-7449 was found to interact with

sequences identical to importin P-subunit (starting at nucleotide 1970, APPENDIX,

Sequence 7), imogen 38 (starting at nucleotide 492, APPENDIX, Sequence 8), DGS-I (a

gene with unknown function, isolated from the DiGeorge syndrome critical region,

starting at nucleotide 455, APPENDIX, Sequence 9), EST cg50136.f6 (IP-1, starting at

nucleotide 1, APPENDIX, Sequence 3), EST AA010799 (IP-2, starting at nucleotide 1,

APPENDIX, Sequence 4), EST cg50136.a7 (IP-4, starting at nucleotide 1, APPENDIX,

Sequence 16), EST cg50175.c7 (IP-5, starting at nucleotide 25, APPENDIX, Sequence

17), EST cg50138.g5 (IP-6, starting at nucleotide 1, APPENDIX, Sequence 18), EST

cg50175.h7 (IP-8, starting at nucleotide 1, APPENDIX, Sequence 20), EST cg50136.a5.b

(IP-10, starting at nucleotide 1, APPENDIX, Sequence 22), and EST cg49432.h3 (IP-12,

starting at nucleotide 1, APPENDIX, Sequence 24). The sequence identical to EST

cg50138.g5 (IP-6, starting at nucleotide 1) interacted also with LYST fragment 10576-








11611. Identified sequences interacting with LYST fragment 9502-10590 were identical

to atrophin-1 (starting at nucleotide 2660, APPENDIX, Sequence 10), embryonic fyn

substrate 2 (Efs2, starting at nucleotide 1665, APPENDIX, Sequence 11), similar to EST

cg50136.clO0 (IP-3, starting at nucleotide 356, APPENDIX, Sequence 5), identical to EST

KIAA0192 (IP-9, starting at nucleotide 5092, APPENDIX, Sequence 21), and similar to

EST cg50138.el.b (IP-13, starting at nucleotide 75, APPENDIX, Sequence 25). The

sequence identical to EST cg50136.clO0 (IP-3) interacted also with LYST fragment 9037-

9585. The sequences identical to casein kinase II P3-subunit (starting at nucleotides 1 and

58) interacted not only with the LYST fragment 3190-4032 in the forward screen, but

also with LYST fragment 10576-11611 in the reverse screen.

In summary, eleven LYST-interacting sequences were found to be identical to

published sequences encoding proteins (14-3-3, B14-3-3, Hrs, calmodulin, casein kinase

II P-subunit, Efs2, importin P3-subunit, estrogen receptor-related protein, imogen 38,

atrophin-1, and DGS-I), two LYST-interacting sequences were found to be highly

homologous or identical to published EST sequences (AA010799 and KIAA0192,

encoding for IP-2 and IP-9, respectively), and eleven LYST-interacting sequences were

found to be novel ESTs (cg50136.f6, cg50136.clO, cg50136.a7, cg50175.c7, cg50138.g5,

cg50173.dlO, cg50175.h7, cg50136.a5.b, cg51287.dlO, cg49432.h3.b, and cg50138.el.b;

encoding for IP-1, IP-3, IP-4, IP-5, IP-6, IP-7, IP-8, IP-10, IP-11, IP-12, and IP-13

respectively).

In the reverse screen searching for LYST2-IPs, one interacting sequence was

found to be identical to the published sequence encoding human brain factor 2 (HBF-G2,








starting at nucleotide 1638, APPENDIX, Sequence 2) and one to the same region

(starting at nucleotide 354, APPENDIX, Sequence 1) of 14-3-3 protein, identified to

interact with LYST. The nucleic acid sequences and corresponding amino acid sequences

of all LYST- and LYST2-interacting proteins are shown in APPENDIX, Sequences 1-25.

A summary of the described protein-protein interactions is shown in Table 5.

Table 5. Protein interactions identified by forward and reverse yeast two-hybrid
screens using fragments of LYST and LYST2 as baits



BAIT FRAGMENT INTERACTING PROTEINS IDENTIFIED IN
AND POSITION,
bp* FORWARD SCREENS REVERSE SCREENS


LYST 190-1056 14-3-3 protein 14-3-3 protein; B14-3-3;

Calmodulin; IP-7; Estrogen
receptor-related protein
LYST 3190-4032 Casein kinase II p-subunit
LYST 4009-4821 14-3-3 protein
LYST 6586-7449 Hepatocyte growth factor Importin P-subunit; Imogen 38;
regulated tyrosine kinase DGS-I (DiGeorge syndrome);
substrate (Hrs) IP- 1 ,-2,-4,-5,-6,-8,- 10,- 12
LYST 9037-9585 IP-1 I IP-3
LYST 9502-10590 Atrophin-1; Embryonic Fyn

substrate 2 (Efs2); IP-3,-9,-13
LYST 10576-11611 Casein kinase II 3-subunit; IP-6
LYST2 774-1424 14-3-3 protein; Human brain

factor-2 (HBF-G2)

* According to GenBank accession number U67615 for LYST; for LYST2, fragment
positions are relative to the beginning of the known sequence









The potential functional significance of the interactions between LYST/LYST2

and the previously published proteins will be discussed in the next chapter. A short

description of all ESTs, detected as LYST-IPs, and their similarity to known sequences

follows.

IP-1 (EST cg50136.f6. APPENDIX. Sequence 13)

One identified sequence (cg50136.f6, 503 nucleotides), interacting with LYST,

was 98% identical to nucleotides 1-128, 90% to nucleotides 339-455, and 71% to

nucleotides 201-445 of the human EST AA452346. This EST of 455 nucleotides was

initially obtained from a human Soares total fetus (8-9 week) library (Hillier et al., 1997).

However, EST AA452346 could be not be used for an extension of EST cg50136.f6 in

either direction.

EST cg50136.f6 showed 68% nucleotide identity to mouse T-complex protein

Tcp-10 genes (GenBank accession No. M73509, M73506, M22601, M73505). An open

reading frame could be translated from nucleotides 3 to 503. The resulting 167 amino

acids correspond to a c-terminal or core region of a novel protein, which was designated

IP-1. Amino acids 2-126 of IP-1 show 68% homology to the c-terminal region (amino

acids 242-365) of mouse Tcp-10 protein (GenBank accession No. X58170)

Therefore, IP-1 represents a novel protein with homology to mouse Tcp-10

protein. Interestingly, another LYST-interactant, IP-10 (see later in this chapter), shows

homology to the amino-terminal region of Tcp-10. In addition, LYST-interactant IP-11

has a region homologous to ATP-binding proteins and chaperonin Tcp-1. The nucleotide

and amino acid sequences of IP-1 are shown in APPENDIX, Sequence 13.








IP-2 (EST AA010799. APPENDIX. Sequence 14)

Three identical sequences (cg50136.a4, cg50136.d6, cg50136.c9), interacting with

LYST, were 96% identical to nucleotides 1-441 of the human EST AA010799. This EST

of 495 nucleotides was initially obtained from a human Soares fetal heart (19 weeks)

library (Hillier et al., 1995). EST AA010799 could not be further extended by database

searches and was used for further analysis.

The 5' end of this EST was similar to the human 50 S ribosomal protein L17. An

open reading frame could be translated from nucleotides 3 to 494 of EST AA0 10799. The

resulting 164 amino acids correspond to a c-terminal or core region of a novel protein,

which was called IP-2. Amino acids 1-107 of IP-2 show 55% homology to amino acids 5-

117 of bacterial ribosomal protein L17 (GenBank accession No. S07223, M26414,

L33834, AE00633, D90905).

Therefore, IP-2 represents a novel human protein with homology to ribosomal

protein L 17. The nucleotide and amino acid sequences of IP-2 are shown in APPENDIX,

Sequence 14.

IP-3 (EST cg50136.clO, APPENDIX, Sequence !15)

Two identical sequences (cg50136.clO and cg50136.glO0), interacting with LYST,

showed significant homology to published ESTs. An extended expressed sequence of

1198 nucleotides was assembled from nucleotides 1 to 466 of human fetal lung EST

W40354 (Hillier et al., 1995), nucleotides 57 to 194 from cg50136.clO (corresponding to

nucleotides 345 to 539 in the extended sequence), nucleotides 1 to 382 from human

endothelial cell EST AA 186481 (Hillier et al., 1995), corresponding to nucleotides 483 to








863 in the extended sequence, nucleotides 336 to 446 from human fetal EST AA436726

(Hillier et al., 1997), corresponding to nucleotides 529 to 974 in the extended sequence,

and nucleotides 199 to 423 from EST N49053 from a human multiple sclerosis library

(Hillier et al., 1995), corresponding to nucleotides 777 to 1198 in the extended sequence.

The fragment interacting with LYST starts at nucleotide 345. The extended sequence of

1198 nucleotides was used for subsequent analysis.

Nucleotides 213 to 599 of the extended expressed sequence showed 75% identity

to nucleotides 3589 to 3975 of rat Olf-1/EBF associated Zn finger protein Roaz

(GenBank accession No. U92564). An open reading frame could be translated from

nucleotides 1 to 594. These 198 amino acids represent a core or c-terminal region of a

novel protein, which was designated IP-3. Amino acids 26 to 197 of IP-3 show 84%

homology to amino acids 1023 to 1185 of the rat Zn finger protein Roaz. Weaker

homologies (40 to 50%) were seen to different zinc-finger proteins of several species.

Interestingly, conserved zinc-finger domains exist in other proteins, e.g., protein kinase C

and Hrs, that may be associated with LYST (see next chapter).

Therefore, IP-3 represents a novel human protein with homologies to rat Olf-

1 /EBF associated Zn finger protein Roaz. The nucleotide and amino acid sequences of IP-

3 are shown in APPENDIX, Sequence 15.

IP-4 (EST cg50136.a7. APPENDIX. Sequence 16)

One identified prey sequence (cg50136.a7, 493 nucleotides), interacting with

LYST, was 91% identical to human EST AA009453, starting at nucleotide 1. This EST

of 505 nucleotides was initially obtained from a human Soares fetal heart (19 weeks)








library (Hillier et al., 1997). EST AA009453 could be extended in the 5' direction with

nucleotides 57 to 76 of clone cg50136.a7, resulting in a sequence of 524 nucleotides.

An open reading frame could be translated from nucleotides 2 to 244. These 81

amino acids correspond to a carboxy-terminal, proline-rich region of a novel protein,

which was called IP-4.

Therefore, IP-4 represents a novel, proline-rich protein without homology to

known proteins. The nucleotide and amino acid sequences of IP-4 are shown in

APPENDIX, Sequence 16.

IP-5 (EST cg50175.c7. APPENDIX. Sequence 17)

One identified prey sequence (cg50175.c7, 548 nucleotides), interacting with

LYST, was 98% identical to human cDNA clone AA403189, starting at nucleotide 24.

This EST of 386 nucleotides was initially obtained from a human Soares total fetus (8-9

week) library (Hillier et al., 1997). EST AA403189 was extended with nucleotides 418 to

548 of clone 50175.c7, resulting in a sequence of 517 nucleotides. This sequence could

not be extended further by database searches and was used for subsequent analysis.

An open reading frame could be translated from nucleotides 1 to 288. This

corresponds to a 96 amino acid protein, which was designated IP-5. Amino acids 10 to 84

of IP-5 were found to be 48% homologous to Bacillus subtilis ATP-dependent protease

La (GenBank accession No. X76424 and P37945), to Bacillus subtilis ATP-dependent

Lon protease (GenBlank accession No. Z75208), and to Bacillus subtilis endopeptidase La

(GenBank accession No. 140421).









Therefore, IP-5 represents a novel protein with homology to ATP-dependent

proteases Lon/La. The nucleotide and amino acid sequences of IP-5 are shown in

APPENDIX, Sequence 17.

IP-6 (EST cg50138.g5. APPENDIX. Sequence 18)

Two identified sequences (cg50138.g5, 498 nucleotides, and the identical

cg50138.f2), interacting with LYST, showed significant homology to published

expressed sequences. Nucleotides 222 to 492 of cg50138.g5 were 98% identical to

nucleotides 1 to 270 of EST AA460131 (588 nucleotides; Hillier et al., 1997). This

sequence was obtained from a human total fetus (9 weeks) cDNA library. A longer

sequence was assembled from nucleotides 1-449 of cg50138.g5 and nucleotides 228-588

of AA460131, and the extended sequence of 753 nucleotides was used for subsequent

analysis.

The extended sequence showed homology to several viral nonstructural proteins:

nucleotides 2 to 487 showed 83% identity to nucleotides 170 to 653 of the nonstructural

protein Ns2-3 of a border disease virus strain (pestivirus type 3) (GenBank accession No.

U43603), nucleotides 42 to 405 showed 90% identity to nucleotides 17 to 378 of proteins

p54, p80 and p125 of the bovine viral diarrhea virus (GenBank accession No. Z54331).

Nucleotides 311 to 753 showed 58% identity to nucleotides 3025 to 3464 of the human

growth hormone and chorionic somatomammotropin genes (GenBank accession No.

J03071).

An open reading frame could be translated from nucleotides 1 to 582. These 194

amino acids represent a core or c-terminal region of a novel protein, which was called IP-









6. Amino acids 6 to 144 of IP-6 show 93% homology to amino acids 64 to 198 of the

viral (pestivirus type 3) nonstructural protein Ns2-3 (GenBank accession No. U43603).

Amino acids 31 to 135 are 98% homologous to amino acids 22 to 126 of the

nonstructural protein p125 of the bovine viral diarrhea virus (GenBank accession No.

Z54332). BLOCKS analysis searching for similarities to known protein families showed

some homology to 14-3-3 proteins (amino acids 17 to 68; identities are shown in capital

letters): Efl SK1QD dLKea mntmm CSRcQ GkhRr Femdr Epksa RycAE cnrlh pAE.

Therefore, IP-6 represents a novel human protein with homology to viral

nonstructural proteins. The nucleotide and amino acid sequences of IP-6 are shown in

APPENDIX, Sequence 18.

IP-7 (EST cg50173.dl 0. APPENDIX. Sequence 19)

One identified sequence (cg50173.dlO, 449 nucleotides), interacting with LYST,

showed homology to published expressed sequences. Nucleotides 79 to 325 of

cg50173.dl0 were 91% identical to nucleotides 30 to 283 of the expressed sequence

H62553 (293 nucleotides; Hillier et al., 1995). This sequence was obtained from a human

male fetal (20 weeks) liver and spleen cDNA library. The expressed sequence

cg50175.dlO could not be further extended by database searches and was used for further

analysis.

An open reading frame could be translated from nucleotides 17 to 289, starting

with a methionine start codon. These 91 amino acids correspond to a novel protein with a

calculated molecular weight of 10589,9, which was designated IP-7. Both the nucleotide

and the amino acid sequence of IP-7 show some homology to Alu-domains.








Therefore, IP-7 represents a novel human protein without homology to known

proteins. The nucleotide and amino acid sequences of IP-7 are shown in APPENDIX,

Sequence 19.

IP-8 (EST cg50175.h7. APPENDIX. Sequence 20)

One identified sequence (cg50175.h7), interacting with LYST, showed significant

homology to published expressed sequences. Nucleotides 254 to 555 of cg50175.h7 were

96% identical to nucleotides 1 to 330 of the expressed sequence T09146. This sequence

was obtained from a human infant brain cDNA library (Adams et al., 1993). Nucleotides

61 to 433 of cg50175.h7 were assembled to nucleotides 181 to 330 of T09146, resulting

in an extended sequence of 523 nucleotides. This sequence could not be further extended

by database searches and was used for further analysis.

Nucleotides 2 to 523 of the assembled expressed sequence were 89% homologous

to the rat mRNA for norbin (GenBank accession No. AB006461; Shinozaki et al., 1997).

Norbin is a novel brain gene, induced by treatment of tetraethylammonium in rat

hippocampal slice and accompanied with neurite-outgrowth in neuro 2a cells. The

neurite-outgrowth-related protein norbin may play a role in neural plasticity because of

the formation of new synapses.

An open reading frame could be translated from nucleotides 2 to 523. These 174

amino acids correspond to a c-terminal or core region of a novel protein, which was

called IP-8 and which was 98% identical to amino acids 486 to 659 of rat norbin brain

protein (729 amino acids).








Therefore, IP-8 represents a novel protein that appears to be the human homolog

of the rat brain gene norbin. The nucleotide and amino acid sequences of IP-8 are shown

in APPENDIX, Sequence 20.

IP-9 (EST KIAA0192. APPENDIX. Sequence 21)

One identified prey sequence of 499 nucleotides interacting with LYST was 96%

identical to human mRNA clone D83783 for the KIAA0192 gene, starting at nucleotide

5092. The mRNA for KIAA0192 was initially obtained from a human male myeloblast

cell line (Nagase et al., 1996).

An open reading frame without a methionine start codon could be translated from

nucleotides 1 to 6372. This sequence represents the carboxy-terminal part of a protein

which was designated IP-9.

Since no homologies to known proteins could be identified, it appears that IP-9,

encoded by the human mRNA for the KIAA0192 gene, represents a novel protein that

interacts with LYST. The nucleotide and amino acid sequences of IP-9 are shown in

APPENDIX, Sequence 21.

IP-10 (EST cg50136.a5.b. APPENDIX. Sequence 22)

Two identified sequences (cg50136a5 and cg50136c4), interacting with LYST,

showed no significant homology to published sequences. Nucleotides 62-309 were 57%

homologous to nucleotides 61 to 310 of the human immunodeficiency virus envelope

protein (GenBank accession No. L08337). The interacting sequence of 451 nucleotides

was used for further analysis.








An open reading frame could be translated from nucleotides 2 to 451. These 150

amino acids correspond to a c-terminal or core region of a novel protein, which was

called IP-10. Amino acids 59 to 139 of IP-10 showed 59% homology to amino acids 29-

109 of the human T-complex protein TCP-10 (GenBank accession No. U03399). Other

homologies (-50%) were found to several cytoskeletal proteins (e.g. myosin I isoform,

kinesin precursor, cytokeratin, caldesmon, NUF1, tropomyosin, neurofilament protein,

kinesin-like protein KIF1, myosin like protein MLP1, troponin T) and to proteins

involved in vesicular transport intracellularr protein transport protein, nuclear fusion

protein Biki, synaptonemal complex protein). BLOCKS analysis showed homologies to

postsynaptic proteins, gas vesicles protein, clusterin proteins, clathrin light chain proteins,

tropomyosin protein and intermediate filament protein.

Therefore, IP-10 represents a novel protein with homologies to the amino-

terminal region of Tcp-10 and with homologies to cytoskeletal and vesicular transport

proteins. It is interesting that the LYST-interactant IP-1 shows homology to the c-

terminal region of Tcp-10. Furthermore, LYST-interactant IP-11 has homologous

domains to ATP-binding proteins and chaperonin Tcp-1. The nucleotide and amino acid

sequences of IP-10 are shown in APPENDIX, Sequence 22.

IP- 11 (EST cg51287.d 10. APPENDIX. Sequence 23)

One identified sequence (cg51287.d10), interacting with LYST, was 90%

identical to nucleotides 1-325 of the human EST AA310287. This EST of 470

nucleotides was initially obtained from a human Jurkat T-cells library (Adams et al.,

1995). The 5' end of EST AA310287 could be extended with nucleotides 154 to 180 of








EST cg51287.dlO, resulting in a sequence of 497 nucleotides, which was used for further

analysis.

An open reading frame could be translated from nucleotides 3 to 497. These 165

amino acids correspond to a c-terminal or core region of a novel protein, which was

designated IP-11. No significant homologies could be found using the BLASTN search

program. Using the BLOCKS analysis program, homologies to ATP-binding proteins and

chaperonins TCP-1 proteins could be found.

Therefore, IP-11 represents a novel protein with homologies to ATP-binding

proteins and TCP-1. The nucleotide and amino acid sequences of IP-11 are shown in

APPENDIX, Sequence 23.

IP-12 (EST cg49432.h3.b. APPENDIX. Sequence 24)

Two identified sequences (cg50136c6 and cg50136d2), interacting with LYST,

showed no significant homologies to published sequences. The interacting sequence of

402 nucleotides was used for further analysis.

Nucleotides 27-266 of EST cg49432.h3.b were 70% homologous to nucleotides

1115 to 1348 of the human RNA for the cellular oncogene c-fes (GenBank accession No.

X52192).

An open reading frame could be translated from nucleotides 2 to 400. These 133

amino acids correspond to a c-terminal or core region of a novel protein, which was

called IP-12. Only weak homologies to known proteins could be found, especially to

proline-rich regions of different proteins. For example, the proline rich region of IP-12

consisting of amino acids 57 to 89 was 69% homologous to the TGF-3 binding protein 3,








amino acids 57 to 75 were 66% homologous to a region in human semaphorin IV, 64%

homologous to StpC139 (Saimiriine herpesvirus 2), 63% homologous to a collagen-like

protein and 55% homologous to homeobox protein HOX-A4. Although the functions of

proline-rich regions are often unclear, they could be involved in protein-protein

interactions. Some transcription factors contain a proline-rich sequence in the

transcriptional activation domain, which has no DNA-binding activity but is essential for

activating transcription (Mitchell and Tjian, 1989). Amino acids 14 to 67 of IP-12

showed 53% homology to ORF homolog membrane peroxisomal 70 kDa protein P34230

(ALDP), involved in adrenoleukodystrophy.

Therefore, IP-12 represents a novel protein with a proline-rich region. Other

LYST interactants, IP-4 and Hrs, have also proline-rich regions. The nucleotide and

amino acid sequences of IP-12 are shown in APPENDIX, Sequence 24.

IP-13 (EST cg50175.cl 1. APPENDIX. Sequence 25)

One identified sequence (cg50175.c 1, 481 nucleotides), interacting with LYST,

showed significant homology to published expressed sequences. Nucleotides 1 to 355 of

cg50175.cl 1 were 96% identical to nucleotides 74 to 428 of the expressed sequence

H51347 (428 nucleotides; Hillier et al., 1995), obtained from a mouse brain cDNA

library. The expressed sequence cg50175.cll was extended in the 5' direction with

nucleotides 1-73 of H51347. This extended sequence showed 96% identity to an

expressed sequence Hs21767 of the same length. The extended sequence of 554

nucleotides was used for further analysis.








The extended expressed sequence of 554 nucleotides showed 71% identity to

nucleotides 463 to 982 of Xenopus laevis elav-type ribonucleoprotein etr-1 (GenBank

accession No. U 16800). An open reading frame could be translated from nucleotides 3 to

554. These 184 amino acids represent the core or c-terminus of a novel protein, which

was designated IP-13. The amino acid sequence of IP-13 shows homologies to

ribonucleoproteins of different species, including 67% homology to Xenopus laevis elav-

type ribonucleoprotein etr-1 (GenBank accession No. U16800) and 54% homology to the

human etr-3 protein (GenBank accession No. U69546).

Therefore, IP-13 represents a novel human protein with homology to

ribonucleoproteins. The nucleotide and amino acid sequences of IP-13 are shown in

APPENDIX, Sequence 25.














DISCUSSION


Since the predicted mouse and human CHS proteins were novel and unlike any of

the molecules previously implicated in vesicular transport, their identification did not

prove immediately helpful in establishing the precise mechanism whereby CHS

dysregulates protein transport and lysosomal trafficking. Therefore, the main goal of this

dissertation was to characterize the Chediak-Higashi gene and its products in more detail

by using functional genomics approaches:

identifying alternatively spliced Lyst mRNA isoforms and therefore resolving

the discrepancy between the initially reported Lyst sequences

identification of mutations in Chediak-Higashi patients and beige alleles

identification of proteins that interact with LYST





Identification and Characterization of Lyst mRNA Isoforms


In order to investigate the expression of the CHS gene and to explain the

discrepancy between the previously published Lyst sequences, extensive northern and

RT-PCR analyses were performed. We demonstrated that each of the previously reported

bg gene sequences (Barbosa et al., 1996; Perou et al., 1996b) is derived from a single

gene with alternatively spliced mRNAs (Fig. 8). The previously reported sequences are









derived from non-overlapping parts of two Lyst mRNA isoforms with different predicted

C-terminal regions. By sequencing RT-PCR products, we have shown that nucleotides 1-

4706 of Lyst also represent the previously undetermined 5' region of the largest Lyst

isoform. Alternative splicing at nucleotide 4706 results in bg gene isoforms that contain

different 3' regions. Splicing from exon a (containing nucleotide 4706) to exon T results

in an -12 kb mRNA (Lyst-I) that corresponds to the largest band observed on northern

blots. Incomplete splicing at nucleotide 4706 results in a 5893 bp cDNA (Lyst-II) that

contains intron-derived sequence at the 3' end. Lyst-II corresponds to a smaller mRNA

observed on northern blots (Barbosa et al., 1996). While several other genes generate an

alternative C-terminus by incomplete splicing (Myers et al., 1995; Sugimoto et al., 1995;

Sygiyama et al., 1996; Zhao and Manlley, 1996; Van De Wetering et al., 1996), the bg

gene is unique in that the predicted structures of the two C-termini are quite different. The

C-terminus of Lyst-I contains a WD-repeat domain that is similar to the 13-subunit of

heterotrimeric G proteins and which may assume a propeller-like secondary structure

(Lambright et al., 1996). This domain is absent in Lyst-II.

Northern blots of human tissues had suggested that transcription of the

homologous human gene, LYST, had a similar complexity to the mouse. We identified

two human ESTs homologous to mouse Lyst and described a mutation in one of these in

a Chediak-Higashi patient (Barbosa et al., 1996). Subsequently, another group published

the cDNA sequence of the largest LYST isoform, and identified mutations in this gene in

two additional patients with CHS (Nagle et al., 1996). We have described the

identification of a second isoform of the human gene. This mRNA encodes a protein of








1531 amino acids that is homologous to mouse Lyst-II. Like the latter, this human mRNA

arises from incomplete splicing and retention of a transcribed intron that encodes the C-

terminus of the predicted LYST protein. The mouse and human codons unique to this

short isoform share 65% amino acid identity. The stop codon, however, is not conserved

precisely between the human and mouse short isoforms. While mouse Lyst-II is predicted

to contain a C-terminal prenylation motif (CYSP), translation of human short isoform is

predicted to terminate 22 codons earlier and to lack this motif.

However, the other predicted structural features of the human short isoform were

conserved with mouse. The most notable of these was a region similar in sequence to

stathmin (amino acids 376-540) (Barbosa et al., 1996). While the mouse and human CHS

genes had an overall amino acid identity of 81%, identity in the stathmin-like domain was

92% and similarity was 99%. Stathmin is a coiled-coil phosphoprotein that regulates

microtubule polymerization in a phosphorylation-dependent manner, and acts as a relay

for intracellular signal transduction (Sobel, 1991; Belmont and Mitchison, 1996;

Marklund et al., 1996). This region of the LYST gene may encode a coiled-coil protein

interaction domain and may regulate microtubule-mediated lysosome trafficking.

Intriguingly, a defect in microtubule dynamics has been documented previously in CHS

(Oliver et al., 1975) and intact microtubules are required for maintenance of lysosomal

morphology and trafficking (Matteoni and Kreis, 1987; Swanson et al., 1987; Swanson et

al., 1992; Oka and Weigel, 1983).

Mutation and expression analyses suggested that the largest Lyst splice variant

(Lyst-I) represents the isoform of primary functional significance and that expression of

the smaller isoform (Lyst-II) alone cannot compensate for loss of Lyst-I. This hypothesis








was supported by the genetic complementation experiments initiated in collaboration

with Dr. Stephen Brandt. These studies revealed that some beige mouse fibroblasts,

expressing full-length Lyst-I, demonstrated more peripheral distribution in the cytoplasm

and slight reduction of the size of the stained granules. Such changes were not visible in

cells expressing the smaller Lyst-II isoform. However, since the effects of the

complementation were not very prominent and were not observed in all cells expressing

Lyst-I, the significance of the described genetic complementation is questionable.

Comparison of the relative abundance of LYST gene transcripts in human tissues

at different developmental stages revealed an overlapping but distinct pattern of

expression (Fig. 9). A quantitative estimate of the relative expression of the smaller

mRNA isoforms was generated by comparing the relative hybridization intensity

obtained with a probe specific for the large isoform with that obtained with a probe that

hybridizes to all LYST transcripts. Large isoform transcripts predominated in thymus,

fetal thymus, spleen and brain (with the exception of amygdala, occipital lobe, putamen

and pituitary gland). Large and small mRNA transcripts were abundant in the latter brain

tissues, peripheral blood leukocytes and bone marrow. However, in several tissues, only

the small isoforms were expressed, e.g. in fetal heart, salivary gland, kidney, lung, and

fetal brain. The developmental pattern of mRNA isoform expression in brain was

particularly interesting, since only the small isoforms were expressed in fetal brain,

whereas the largest isoform predominated in many regions of the adult brain.








Identification of Mutations in Patients with Chediak-Higashi Syndrome and in Beige

Mice


Mutation analysis was performed in order to provide additional evidence that

LYST, resp. Lyst is indeed the gene responsible for the CHS in human and for the beige

mutation in mouse, as well as to determine whether a genotype-phenotype correlation

could be established between different genetic defects and specific disease

manifestations. Novel mutations were identified within the coding domain of LYST in

several CHS patients (Table 4 and Fig.3). The genetic lesions in three patients (370, 372,

373) were C--T substitutions that resulted in premature termination. Another patient was

heterozygous for a dinucleotide deletion that results in a frameshift and premature

termination. Interestingly, all bg and LYST mutations identified to date are predicted to

result in the production of either truncated or absent proteins, suggesting that missense

mutations may not be likely to cause CHS. Unlike Fanconi anemia, type C (Yamashita et

al., 1996) there does not appear to be a correlation between the length of the truncated

proteins (which may or may not be stable) with clinical features or disease severity in

CHS patients, given that a CHS patient with mild manifestations has a truncating

mutation early in the protein. However, until the other mutant allele in the compound

heterozygote patients is identified, and the exact effects of each mutation at the protein

level are characterized, such correlation is imprecise.

Mutation and expression analyses were also useful in addressing the important

question about the biological relevance of the 2 main LYST transcripts in regulating

lysosomal trafficking and in preventing Chediak-Higashi syndrome. The distribution of