Cell lineage mutants in the nematode Caenorhabditis elegans

Cell lineage mutants in the nematode Caenorhabditis elegans

288 I'IN,~ b o u n d 'extrinsic' g u i d a n c e c u e s m i g h t n o t be n e c e s s a r y E v e n if intrinsic c u e s w e r e n o t sufficientl...

502KB Sizes 1 Downloads 41 Views

288

I'IN,~

b o u n d 'extrinsic' g u i d a n c e c u e s m i g h t n o t be n e c e s s a r y E v e n if intrinsic c u e s w e r e n o t sufficiently p r e c i s e to g u i d e e x t e n d e d o u t g r o w t h , t h e y c o u l d act in h a r m o n y with e x t n n s l c s u b s t r a t a cues, w h i c h w o u l d be u s e d to c o r r e c t s m a l l e r r o r s or reduce large c h a n g e s m the g r o w i n g a x o n s ' s trajectory_ T h i s h y p o thesis r e m a i n s , h o w e v e r , u n t e s t e d

Conclusions It is possible, a n d p e r h a p s e v e n h k e l y , that a x o n o u t g r o w t h t h r o u g h the fly wing ts d i r e c t e d by a v a r i e t y o f mechanisms; such redundancy would certainly h e l p to e x p l a i n t h e s t e r e o t y p y observed during normal nerve formatton a n d t h e r e s i s t a n c e to p e r t u r b a t i o n observed m experimental studies Thus, even though both neuronal cues a n d closed c h a n n e l s c a n b e e l i m i n a t e d f r o m t h e w i n g (singly or s i m u l t a n e o u s ly) w i t h o u t p r e v e n t i n g t h e f o r m a t i o n o f correctly p o s i t i o n e d s e n s o r y n e r v e s , t h e s e factors m a y still play a role in normal axon guidance Our experim e n t s do d e m o n s t r a t e , h o w e v e r , t h e e x i s t e n c e o f o t h e r t y p e s of a x o n g u i d a n c e c u e s in t h e wing, a n d s e v e r a l

s t u d i e s are u n d e r w a y m a n a t t e m p t to test a n d c h a r a c t e r i z e t h e s e cues.

Selected references 1 Johnston, R N and Wassails, N K (1980) in Curr Top Dev B:ol 16, 165-206 2 Letourneau, P C (1982) m Neuronal Development (Spltzer, N C , ed ), pp 213-254, Plenum Press, New York 3 Gunderson, R W and Barrett, J N (1980) J (ell B:ol 87,546-554 4 Patel, N and Poo, M-M (1982) J Neuroscl 2, 483--496 5 Weiss, P (1961) Exp Cell Res Suppl 8, 26(I--281 6 Lew-Montalclni, R , Chen, J S,, Seshan, K R and Aloe, L (1973) in Developmental Biology of Arthropods (Young, D., ed ), pp 5-36, Cambndge University Press, Cambridge 7 Soloman, F (1979) Cell 16, 165-169 8 Waddmgton, C H (1940)J Genet 41, 75139 9 Palka, J , Schublger, M and Elhson, R (1983) Dev B:ol 98, 481--492 10 Jan. L Y and Jan, Y N (1982)Proc Nail Acad Sci USA 79, 27(]0-2704 11 Murray, M_ A , Schulmger, M and Palka, J (1984) Day Blol 104. 259-273 12 Jan, Y N , Gbysen, A and Jan, L Y (1983) Soc Neurosci Abstr 9, 302 3 13 Bate, C M (1978) m Development of Semory Systems (Jacobson, M , ed ), pp 1-

Cell

lune 1985

53, Sprmger-Verlag, New 'fork 14 Goodman, C S, Bastmnl, M J , Doe. C O , d u L a c . S , Helfand, S L ,Kuwada, J Y and Thomas, J B (lq84) Sc+ence225, 1271-1279 15 Bentley, D and Keshtshtan+ H (1982) Trends NeuroSct 5,354-'~58 16 Bentley, D and Caudy, M (1983) Cold Spring Harbor Syrup Quant Btol 48, 57.~585 17 Marnn, P and Schneider, [ (1978) m The Genencs and Bzology o] Drosophila (Ashburner, M and Wnght, T R F , e d s ) pp 219-264, Academic Press, London 18 Siegel, J G and Fnstrom, J W (1978) lb~d, pp 317-394 19 Edwards, J S, Mdner, M J and Chen, S W (1978) Wilhelm RoutAreh 185, 59-77 20 Blair, S S and Palka J De+ Biol (m press) 21 Schublger, M and Palka, J Day Blol (m press) 22 Anderson, H (1984) Wilhelm Rout Arch 193,226-233 23 Blmr, S S, Murray, M A and Palka, J Nature (in press) 24 Nardt. J B (1983)Day B*ol 95, 163-174 25 Berlot, J and Goodman, C S (1984) Science 223,493-496 Seth S Blmr and John Palka are at the Department of Zoology, Umvemty of Washington, Seattle, WA 98195, USA

in ttte Edward M. Hedgecock

N e m a t o d e s d e v e l o p by m v a n a n t cell hneages. In m a n y cell hneage mutants, spec:fic p r e c u r s o r cells duplicate subhneages n o r m a l l y m a d e elsewhere in the a m m a l . F r o m such mutants, we can infer h o w dtfferent cell types m a y be specified in n o r m a l d e v e l o p m e n t . E v e r y i n d i v i d u a l h a s a u n i q u e , if u n r e c o r d e d , cell h n e a g e d e s c r i b i n g all cell divisions f r o m z y g o t e to adult. In the small nematode, Caenorhabdl~ elegans, t h e s e q u e n c e of cell divmions, a n d t h e f a t e s o f the cells g e n e r a t e d , a r e e s s e n t i a l l y m v a r i a n t b e t w e e n indiv i d u a l s 1-5 In h e r m a p h r o d i t e s , for e x a m p l e , precisely 1076 cell divisions o c c u r , c r e a t i n g 1076 s o m a t i c cells plus t h e g e r m l i n e p r e c u r s o r cell The p a t t e r n o f divisions is c o m p l e x A s 2 i ° = 1024, j u s t o v e r t e n division r o u n d s w o u l d s e e m e n o u g h to g e n e r ate all o f t h e s o m a h c cells In fact, s o m e t e r m i n a l cells a n s e after as few as e i g h t divisions w h e r e a s o t h e r s arise a f t e r as m a n y as e i g h t e e n divisions T h e o r g a n i s m is also c o m p l e x D e s p i t e its s m a l l cell n u m b e r , o v e r two h u n d r e d t e r m i n a l cell types, a b o u t half of w h i c h a r e n e u r o n a l , can be distln-

gulshed3.6, 7 , T h e goal of cell h n e a g e s t u d i e s is to u n d e r s t a n d h o w d i v e r s e cell t y p e s are m a d e , c o n c o m i t a n t with i n c r e a s i n g cell number durmg development It is b e h e v e d t h a t t h e fates o f m a n y cells m t h e n e m a t o d e are specified t h r o u g h a sequence of somatically heritable choices E a c h choice c o m m i t s t h e cell a n d its d e s c e n d a n t s to a s u b p a t h o f d e v e l o p m e n t a n d r e s t n c t s it f r o m some immediate alternative Much of d e v e l o p m e n t a l biology is c o n c e r n e d w]th h o w a n d w h e n t h e s e c h o i c e s , sometimes called determinative s w t t c h e s , are m a d e A l t h o u g h the molecular mechanisms underlying the switches are yet u n k n o w n , t h e c o m p o n e n t s c a n be identified by m u t a t i o n s that alter cell l i n e a g e s a n d fates T h e precise c o r r e l a t i o n b e t w e e n cell a n c e s t r y a n d s u b s e q u e n t fate m C

(~ 1985,ElsevierSoence PubhshersB V Amsterdam037~ 5912./85/$f)2LI1

elegans s u g g e s t s that fate specification p r o c e e d s a p a c e with cell division in this o r g a n i s m . W i t h s o m e exceptaons, the fates o f i n t e r m e d i a t e cells m t h e l i n e a g e are n o t a l t e r e d by e x p e r i mental ablations of neighboring cells 1,3,4,s-1° T h i s s t r o n g l y suggests t h a t i n t e r m e d i a t e cells, t h o u g h l a c k i n g m u c h o f t h e o v e r t d i f f e r e n t i a t i o n of t e r m i n a l cells, a r e irreversibly c o m m i t t e d to g e n e r a t i n g a specific s u b h n e age a n d set o f d e s c e n d a n t s

Mother and d a u g h t e r s E a c h cell division involves t h r e e cells, a m o t h e r a n d two d a u g h t e r s . In c u l t u r e d cell lines, all t h r e e cells are u s u a l l y of the s a m e type I n d e e d , t h e n o r m a l p r o c e s s of cell c o m m i t m e n t , and controls on proliferation, may h a v e to be c i r c u m v e n t e d to o b t a i n i m m o r t a l cell lines o f u n i f o r m cell type In C elegans, m o t h e r a n d d a u g h t e r cells g e n e r a l l y h a v e t h r e e distract fates (Fig 1) T h u s , a m e c h a n ism, p r o b a b l y o p e r a t i n g at t h e t i m e o f cell division, ts r e q u i r e d to d i s t i n g u i s h

289

TINS-June1985

A

A II

IIAII

1

............I ............

I

I

B' WILD TYPE

I ......

i Ic

lin-17

I'

UAII

unc-86 Fig. 1. In a typical cell &vision m the wdd-type, the two daughter cells (B and C) have different fates In s o m e cell divisions m hn-17 and unc-86 mutants, one daughter fads to generate Its normal hneage but generates a hneage charactenstw o f its sister or Its mother, respectively 11 ~2 In the unc-86 case, the error ts c o m p o u n d e d at each dtvtston, gtvmg the hneage mdefimte length lt ts not k n o w n whether two unc-86h k e switches ever act concomitantly, one m each daughter, or tf a hn-17-hke gene acts with the unc-86 gene to determine both daughters o f a single division

mother from daughters and sister from sister In hn-17 and unc-86 mutants, the mechanism of determination fails in specific cell divisions and one daughter pursues an abnormal fate In hn-17 mutants, the abnormal daughter adopts the fate of ~ts normal sister 1~ In consequence, an asymmetrical lineage becomes symmetrical (Fig 1) In unc86 mutants, the abnormal daughter repeats the fate of the mother rather than proceeding to its own fate t2 The pattern of division and differentiation that results is what is known as a stem cell hneage (Fig 1). Simple switching models that can account for such mutants are shown in Fig 2 The axes and polarity of cell divisions, defined by the different fates of the daughters, are relatively lnvanant m nematodes The great majority of cell divisions are aligned, at least approximately, along the anterior/ posterior axis of the ammal but both left/right and dorsal/ventral cell divisions also occur 1-4 Dividing cells may be intrinsically polar and asymmetric cell contacts may merely orient them. Alternatively, asymmetric contacts may actually create the polarity Reversals of d~vislon polanty have been observed in bypodermal and gonadal hneages when neighbors are removed 8 ,) Interestingly, a uterine precursor may divide without polarity, duphcatmg one daughter fate as m hn-

17 mutants, when its nearby sister is removed l0

The first division of the egg The first embryonic division is especially attractwe for studying the mechanism of fate specification The zygote is a large cell and division events are not complicated by neighboring cells. As in most divisions m C elegans, the daughter cells are not equivalent 4 The anterior daughter (with the classical name AB) generates most of the ectoderm (hypodermls, nervous system) The posterior daughter (with the classical name PI) generates most of the mesoderm

a.

The radial conglomeration of mlcrotubles formed around the centnoles dunng cell djws]ons

*

C.

b. A(O)

B(I)

(muscles, glands, gonad), the endoderm (intestme), and the germhne The partition is not absolute as AB generates some muscles and P1 generates some neurons and part of the hypodermls A B is about 50% larger in volume than P1- The immediate cause of the asymmetric cleavage is a posterior migration of the P1 aster* during anaphase 13 As some cell divisions gwe daughters of equal size, but noneqmvalent fates, the mechanism of determinatlon cannot rely In general on differences in cytoplasmic volume Rather, the eccentric positioning of the asters may rely on the same underlying asymmetry used in determination Another zygotic asymmetry anses even earlier Cytoplasmic particles, called P-granules, dispersed throughout the oocyte, move to the posterior periphery of the cell following fertilization I4 By the first cleavage, they reside almost exclusively In P1- This remarkable movement is repeated dunng prophase of each subsequent division until, four divisions from the zygote, the P-granules are found only in the cytoplasm of a single cell, the germhne precursor P4 Although their function is unknown, the P-granules show that a mechanism exists that can segregate cytoplasmic components to only one daughter through consecutive cell divisions What cytoskeletal mechamsms produce these asymmetries9 An asymmetnc cleavage could occur if the two centrosomes differ m their capacity to nucleate mlcrotubules, or alterna-

A(O0)

C(21

B(OI)

C(ll)

A(OO

I

1

Fig. 2. A ternary switch (a) or a combmatton o f two binary switches tb and c) would suffice to dtsangutsh three cell types M o d e l (b) can accommodate the hn-17 phenotype by assuming the switch represented by the first digit ts locked tn the O posttzon It can also accommodate unc-86 if the second switch ts locked m the 0 posttton and the normally unreahzed state 10 ts interpreted as a cell o f type C M o d e l (c) can accommodate unc-86 by assuming the second swttch ts locked tn the 0 position It can accommodate hn-17 only If this same switch can also lock tn the I position and the normally unreahzed state 11 ts interpreted as a cell o f type B

290

fINS

metric pos~tiomng of the asters m anaphase Th~s suggests that a contractile system of actm fdaments may be responsible for some or all of the asymmetries of the first dw~ston

ZYGOTE

I

!

l

P1

AB

I

A Bo

I

A Bp

/\

~

I /~ L~R

E0

/=~R

I Ep L~R /~ L = R

I

I /~ L = R

~Ig. 3. The early dwts~ons of the zygote generate aght cells (ABa, ABp, MS, Ea, Ep, C, D, P4) Stx of these cells chwde into ~denacal (Ea, Ep and D) or nearly tdenncal (.481), MS and C) lefilnght daughters ¢ Bdaterally symmetric dtws~ons do not occur m the ABa hneage before ~tsfourth round of dwtston The cell P~ founds the germ hne The prmetpal blast cells have been gwen h~stoncal names in uppercase letters ~ For each subsequent dw~s~on, a lowercase letter ~s appended reflecting the posmon of the cell relatwe to its s~ster For example, ABpl ts the left daughter of the posterior daughter of the blast cell called A B whtch aself zs the anterior daughter of the zygote

twely, the posterior cell cortex has a reduced abihty to capture and stabilize the ends of astral microtubules ia The movement of the P-granules depends on actin rmcrofilaments but not on cytoplasmic Imcrotubules or the mitotic spindle I4 Cytochalasins which dts-

June 1985

rapt microfilaments block the m~graaon of the P-granules and cause them to aggregate in the center of the cell These drugs also block the waves of furrowing that pass over the anterior cortex of the zygote concurrently wsth P-granule migration, and the asym-

Symmetry and uniqueness A simple strategy for creating an organism with bilateral symmetry is to form identical left/right daughters at the first dlvis]on. This hneage makes no allowance for unique, midplane cells All ttssues must etther comprise identical contributions from the left and the nght precursor cells or violate the bdateral symmetry already established An alternatwe strategy is to set aside cells for unique structures before the lineage undergoes bilaterally symmetrical dwisions Examples of both strategies, symmetricaI division with later violations to create unique cells and prior segregatton of umque cells, occur m the C elegan~ hneage (Fig 3) The first lineage intermediate to undergo asymrnetnc division ~s ABp (lineal names are explained in Fig. 3 legend) Most, but not all, ABpl decendants have fates ~dentical to their ABpr homologs (Fig 4) Among the

~D ~r

I

I

I I

l

t'-] C: "UX

3

4

:i )X

V-

F~g. 4. ComparLson o f the lineages of the left and right daughters of A B p (Re.f 4) Cells that &fief from thmr homologs are shown m boldface type The fate o f the left homolog Is shown first, separated from the fate of the right homolog by a slash For pmrs m which the fates of the homologs d~ffer between mdw~duals, P1/P2, PgP~, P~/P6, PT/Ps, Po/PI~ PH/PIz, hypS/hyp9 and DBI/DB3 pairs, the anterior fate ~ shown first. Programmed cell deaths are marked by crosses The dtvtston axes are antertor (a)/postenor (p) unless marked (d, dorsal, v, ventral, l, left, r, ngh 0 The anterior daughters are drawn to the left

TINS-June

291

1985

exceptions are cells that intercalate with their contralateral homologs at the ventral mldline where they jockey for position and fate For example, either ABplapappa or ABprapappa can occupy the most posterior (P12) position in the queue of ventral ectoblasts while its homolog assumes the penultimate ( P l l ) position The subsequent hneages generated by P l l and P12 are different Thus, the ABpl/rapappa homologs are equivalent cells directed to alternative fates by cell interactions that occur m the queuing process. If either homolog is kdled before entenng the queue, the surviving homolog invarmbly assumes the P12 position and fate 8. The P12 fate is called primary and the P l l fate secondary in the hierarchy defined by experimental ablation and replacement Some cell eqmvalences are only revealed by ablation of the cell which normally assumes the primary fate For example, ABplapaapa consistently assumes the G2 fate, leaving the W fate to its homoiog, ABprapaapa If either cell is killed, however, the survivor assumes the primary (G2) fate 4 The homologous pairs with fates 1/2, 3/4, 5/6 and 7/8 (Fig 4) constitute a second class of differences between the ABpl and A B p r lineages_ They are embryonic intermedmtes rather than terminal cells, and apparently cannot replace each other. How might homologous cells on the left and right sides be instructed to adopt distract fates9 An extnnslc signal, localized to one side of the embryo, could divert specific ceils on that side to alternative fates. A binary switch, turned ON in one A B p daughter and inherited by all its descendants, could accomplish the same purpose In h n - 1 2 mutants, certain left/nght homologs with normally different fates adopt the same fate 15_ For example, the ABpl/rapaapa homologs both adopt the secondary (W) fate in recessive mutants with no l i n - 1 2 function These same cells both adopt the pnmary (G2) fate m dominant mutants believed to have an overactive or non-repressible h n - 1 2 product. Although h n - 1 2 affects several different pairs of homologs descending from ABp, many left/right differences, including the early divergences, remain in these mutants This suggests that I m - 1 2 itself is not a part of a regionspecific signal or a heritable early switch but acts further downstream Indeed, some ABpl descendants ap-

I

A

I

I

A I

I

B

I

O0 A

I

I

OI B

I

I

C

A

I

I I B

O0

O0

I

I

I

I0 C

!

!

II D

I

O0 A

I

I

C

I

I0 C

I

I III

OI B

D

Fig. 5. Mode& for generating :denttcal sets of cells by non-tdenl~cal hneages In model (a), determinants are d~trtbuted asymmetrtcally m the cytoplasm Differences m the plane of dtvts:on alter the segregatton of these determinants to the daughter cells (modtfied from Ref 16) In model (b), two binary switches are thrown m interchanged order When grandaughiers are etammed, the fates of sisters and cousins are interchanged

pear to require the h n - 1 2 product O F F (e.g DA9) while others require it ON (e g G2) Different paths to the same place Just as bilateral homologs do not always have analogous fates, bilateral analogs do not always have homologous lineal ongms For example, the lineage generated by ABplaaa (fate 1 in Fig 4) on the left side of the animal has a nearly exact replica on the right side Surpnsmgly, the analogous precursor on the nght IS ABarpap, not the homologous cell ABpraaa This might he explained if position rather than lineage determined the fates of these intermediates Alternatively, it suggests that two different switching histories can produce equivalent cells There are many other instances where cells of equivalent potential can he generated by two or more alternative hneages 2-4,i6.i7_ The s i n e q u a n o n is that the sister cell fate is different in each alternative In some cases, the sister cell fate in one lineage is

transferred to a cousin cell or more distant relation in an alternative lineage These shuffhngs of the hneage, called 'phase shifts' or 'altered segregations', may reveal something of its mechanism 3,4,16A7 A model, involving altered partitioning of cytoplasmic determinants, explains how identical sets of cells could be generated by nonidentical lineages (Fig 5a) A simple switching mechanism that could accomphsh the same purpose IS a sequence of two binary switches that are thrown in interchanged order (Fig 5b) In actual instances of altered segregation, the sets of descendants overlap but are not identical Programmed cell death During development of the hermaphrodite, 131 cells undergo programmed death TM Most die within an hour of their birth, often without overt differentiation These unnamed cells are marked only by crosses in the lineage tree (Figs 4 and 5) In a few cases, cells differentiate and serve a

2q2

FIN ~ .lune 1985

function before they die For example, the embryonic tad spike cells hve for about six hours and probably function m shaping the tad (cells ABpl/rppppppa m Fig 4). In other cases, an underlying fate can be inferred from homologs that survive tn other parts of the animal or m the opposite sex For example, two embryonic cells that undergo programmed cell death m males survive in hermaphrodites as HSN neurons required for egg-laying (cells ABpl/rapppappa m Fig 4) In ced-3 and ced-4 mutants, cells that normally undergo programmed cell death instead survwe and, in some mstances, dffferentmte into recogmzable cell types (H Ellis, personal communication, and Refs 11 and 18) For example, m these mutants V5.paapp can d~fferentmte into a P D E neuron, the normal cell type of its aunt (F~g. 6) Some exhumed cells may be m switching states that do not normally occur, and so are not decoded into known cell types The function of cell death m nematodes appears to be to remove unneeded cells Cells which d~e immedmtely after b~rth are thought to be merely by-products of the lineages which generate their surviving sisters There are seven separate instances m

the embryonic hneage where the aunt, as well as the sister, of a surwvmg cell dies (c g ABpl/rapappppa in Fig 4)_ Apparently, the series of choices that specify these cell types are not easdy compressed into a single cell division Sex and sexual dimorphism Sex in C elegans is determined by the karyotype Animals with 1 X chromosome are males, ammals with 2 X chromosomes are hermaphrodites, l e females that produce sperm for self-fertlhzauon Sexual dimorphlsm arises m four ways First, two classes of neurons (CEM and HSN) undergo sex-specific cell death m the embryo 4 Second, certain postembryomc precursors (B, F, U and Y) dwide only m males x. Third, other postembryomc precursors have different lineages m the two sexes I-3 Finally, the germhne of males produces only sperm whereas the germllne of hermaphrodites inltrolly produces sperm and then sw~tches to oocyte produchon 9 Seven known genes are revolved m d e t e r m i n i n g s e x 19~j From studies of double mutants, these genes have been placed m an ordered pathway that measures the X to autosome ratio and uses that ,nformat~on to control the final product, tra-l_ T h e tra-I

actiwty, normally expressed only m XX ammals, directs precursor cells in sex-speofic lineages along the female alternatwe XX ammals that have no tra-1 actwlty develop as males Conversely, XO ammals that have gratuitous tra-I act~vlty develop as females Spatial and temporal patterns Nematodes, though unsegmented, have some translat,onal symmetries. For example, a row of ten hypodermal precursors (H0, H I , H2, V1-V6 and T) on each side of the animal generate neurons, sensilla support cells, and more hypodermal cells during larval development (Fig 6) Their hneages can be described as elaborations on the theme of a hypothetical stem cell which divides once m each larval stage, producing an anterior daughter which fuses with the hypodermal syncytmm (hyp7) and a posterior daughter which retains the stem character In fact, seven different patterns occur m the male Moreover, the dlwslon patterns within each larval stage are distinct Mutations m the ha-20 and hn-22 genes transform the lineages of anterior precursors into patterns normally generated by more posterior homo-

QV5 m --~HO

H1

H2

I

Vl-V4

g

g

?

I

I

2

Rn

o

Fig. 6. H. V ond T hneages of the wdd-type male J Ten pcecarsors, ca!led seam cells, form a row from head to tall along each sate of the animal. The cells dlv:de during the four larval stages to generate neurons (PDE, PHC, PLN, PQR, PVD, PVIt, PVN, PVW, RnA, RnB and $DQ), sensory support cells (ADEso, PDEsh, PDEso, PHsol, PHso2 and RnsO, and three types of hypodermal cells (closed ctrcles, cells whtch fuse wEththe hyp7 syncytmmj open circles, seam cells; andhalf-closedclrctes, tadseamcells) Thetenprecursors, andcertamhneageintermediates, arethemselvesseamcellsandfuncaonmmakingareglonof cuttcle The H2 and T cells also functton as socket cells for the delrM and phasmut sensdla, respecuvely, m newly hatched larvae Th~ funcnon ls segregated to the descendants labelled ADEso and PH~o, respecavely, m the first larval stage The four larval stages (LI, L2, L3, L4) and the adult stage are separated by molts (arrows)

T I N S - June 1985 logs_ In hn-20 mutants, the H1 precursor generates a lineage identical to H2 (unpubhshed) In hn-22 mutants, precursors V1-V4 generate lineages essentmlly identical to V5 (W Flxen, personal communication and Ref. 22) In consequence, these animals have ectoplc postdeind and ray sensilla spaced along the body Heterochronlc mutations transform the hneages generated in certain larval stages into the patterns normally generated at earher or later larval stages1223 In precocious hn-14 mutants (recessive alleles), the first stage division pattern is skipped and the second, third and fourth stage patterns are generated one stage earlier than normal 23 In retarded hn-14 mutants (dominant alleles), the first stage pattern occurs normally and IS then essentially repeated in every subsequent larval stage The lineages in the retarded mutants do not terminate after the normal four larval stages, but continue, punctuated by supernumerary molts, in the adult Proliferation and commitment

Lineages in C elegans, excepting the germhne, are of definite length During larval growth, cells divide at precise times after periods of quiescence Some instructions on the timing and the limits of cell division must be mtnnslc to committed precursors Cell death is probably not used to truncate hneages as none of the cells spared m ced-3 mutants are known to divide again (H_ Ellis, personal communication) Inductive signals from unrelated but adjacent cells coordinate the position and t~mmg of some postembryonic events 3,9,I°. In organisms, like humans, with many more cells than C. elegans but a relatively modest increase m cell types, many cell dwlsions occur without changing the commitments of one or both daughters The extents of these lineages are controlled by extrinsic signals Interestingly, the adults of larger nematodes, such as Ascans, have several orders of magnitude more cells in most tissues than C_ elegans (The nervous system IS a notable exception ) The hneages that generate

293 these extra cells are unknown but, as embryogenesis is fairly conservative in nematodes 4, presumably they represent extended divisions of the postembryonic precursors found in C elegans or postembryonic divisions of cells which are non-dividing in C

ently disparate lineages There are formal similarities between switches which distinguish lineal, spatial, temporal, and sexual homologs They may share partly common molecular mechanisms

elegans 5

Selected references

Conclusion

If cell fates are specified through sequences of heritable choices made during development, there are three basic problems to solve tn each organism: (1) the mechanisms of the individual switches, including when they occur, whether they mvolve signals from other cells, and their mode of somatic inheritance, (2) the controlling logic by which the next switch is selected given the past switching history of the cell, and (3) the decoding rules that map the final switching states to terminal cell fates_ The wild-type nematode lineage is informative because it makes many different cell types, and equally important, because many cell types are made m several copies Equivalent terminal cells can arise by different lineages and can have very different relatives These observations are inconsistent with a popular developmental model, sequential partitioning of potential fates among descendants 24. The map from switching states, the heritable record of cell commitments, to differentlated cell types must be more general Probably somewhat different switching states can yield cells of the same type and very similar states can yield radically different types Whereas terminal cells are grouped Into classes based on their final differentiation, intermediate cells are inferred to be equivalent in commitment ff they either generate identical sublineages or, in cell ablation experiments, can replace one another Determinative switches are hypothesized mechamsms that distinguish otherwise equivalent cells, whether intermediate or terminal. Homeotlc mutations, thought to disrupt components of such switches, can reveal underlying cell equivalences and indicate possible common steps in appar-

1 Sulston, J E and Horvltz, H R (1977) Dev Btol 56, 110-156 2 Klmble, J and Hlrsh, D (1979) Dev Btol 70, 396-417 3 Sulston, J E , Albertson, D G and Thomson, J N (1980) Dev Blol 78, 542-576 4 Sulston, J E , Schlerenberg, E . White, J G and Thomson, J N (1983)Dev Biol 100, 64-119 5 Hedgecock, E M and White, J G (1985) Dev Biol 107, 128-133 6 Albertson, D G and Thomson. J N (1976) Phdos Trans Roy Soc London B 275, 299-325 7 White, J G , Southgate, E , Thomson, J N and Brenner, S Phtlos Trans Roy Soc London B (in press) 8 Sulston, J E and White, J G (1980) Dev Btol 78, 577-597 9 Klmble, J and White, J G (1981) Dev Btol 81,208-219 10 Kimble, J (1981) Dev B:ol 87, 286-300 11 Sternberg, P W and Horvltz, H R (1984) Annu Rev Genet 18, 489-524 12 Chalfae, M , Horvltz, H R and Sulston, J E (1981) Cell 24, 59-69 13 Albertson, D G (1984)Dev BIol 101.6172 14 Strome, S and Wood, W B (1983) Cell35, 15-25 15 Greenwald, I S , Sternberg, P W and Horvltz. H R (1983) Cell 34, 435--444 16 Sternberg, P W and Horvltz, H R (1981) Dev Bzol 88, 147-166 17 Steinberg, P W and Horvitz, H R (1982) Dev Bzol 93, 181-205 18 Horvltz, H R , Ellis, H M and Sternberg, P W (1982)Neuroscz Commun 1, 56--65 19 Hodgkm, J (1980) GeneUcs 96, 649--664 20 Klmble, J , Edgar, L and Hlrsh, D (1984) Dev Bzol 105,234-239 21 Donlach, T and Hodgkln, J (1984) Dev Btol 106, 223-235 22 Horvltz, H R , Sternberg, P W , Greenwald, I S , Flxsen, W and Elhs, H R (1983) CoM Sprmg Harbor Syrup Quant BIol 48, 453-463 23 Ambros, V and Horvltz. H R (1984) Science 226, 409-416 24 Stent. G S (1985) Cell 40, 1-2

Edward M Hedgecock ts m the Department of Cell Biology, Roche Institute of Molecular Biology, Nutley, NJ 07110, USA