The orthologue of KN1 in Arabidopsis is believed to be SHOOTMERISTEMLESS (STM) defined by mutations that result in plants defective in SAM initiation and maintenance. It is expressed in large quantities in the SAM, like KN1, and sequence data was used to help confirm this hypothesis. Since then the sequencing of the Arabidopsis genome has revealed 8 KNOX1 genes including 4 class1 genes STM KNAT1 KNAT2 and KNAT6. Work on other plant species has revealed KNOX1 genes wherever they have been sought.
KNOX genes (and transcripts) are expressed in overlapping sections of the meristem, but are not found in leaf primordia, hinting at possible regulation and hierarchy. Loss of function mutants revealed that some KNOX1 genes are required for maintenance and proliferation of the SAM, e.g. stm mutants in Arabidopsis. Ectopic expression of KNOX1 genes (gain of function) results in dramatic transformations of leaf tissue as seen in kn1 and Hooded suggesting that the inhibition of KNOX1 genes in leaf primordia is important for correct lateral organ development.
Regulation
Much work has been done on the organisation of the KNOX1 pathways in plants an important insight has been gained by the work done by Ori and Byrne (2000) on the negative regulation of KNOX1 genes in Arabidopsis. Down regulation of the KNOX1 genes in the leaf primordia appear to be closely related to myb-domain transcription factors, specifically ASYMMETRIC 1 and ASYMMETRIC 2. Asymmetric 1 (as1) and asymmetric 2 (as2) display phenotypes with lobed leaves with ectopic meristems and pattern changes in the proximal to distal plane. These are typical of ectopic KNOX1 expression, confirmed by upregulation of KNAT1 and KNAT2 in the leaves of these mutants and a matching mutant phenotype with a transgenic Cauliflower Mosaic Virus 35S promoter KNAT1 construct . However as1 and as2 both had no KNOX expression in the leaf founder cells suggesting that AS1 and AS2 are capable of maintaining KNOX1 genes in an off position in the leaf primordia but another undiscovered factor is responsible for the initial downregulation of KNAT1 and KNAT2.
AS1 and AS2 are closely related genes and share a lot of sequence similarity with their myb-domain orthologues. Byrne et al (2000) was able to create a fine map of the AS1 gene and found it to map directly to a previously isolated orthologue (AtPHAN) of ROUGH SHEATH 2 (RS2) in maize and PHANTASTICA (PHAN) in Antirrhinum. The orthologues are also responsible for negatively regulating KNOX1 genes as well as being expressed in the leaf primordia.
Although clearly they are not able to downregulate all KNOX genes as STM in Arabidopsis and KN1 in maize are still present in the SAM of the wild type. Ori and Byrne furthered their study by showing AS1 and STM were expressed in mutually exclusive parts of the SAM and by isolating stm mutants it was found that AS1 was present in the SAM suggesting that STM is responsible for AS1 regulation. stm mutants are rescued by an as1 double mutation and can produce vegetative meristems but have incomplete floral development suggesting that the redundant gene(s) for STM cannot account for all of its roles in plant development.
It has been shown that KNAT1 is responsible for the redundancy of the stm as1 mutants, this is not surprising since STM and KNAT1 are the more closely related of the class 1 KNOX1 genes in Arabidopsis sharing 44% overall similarity with 70% in the homeodomain. The evolutionary implications of this are interesting. Presumably both genes arose from a duplication but whether the ancestral gene had AS1 inhibition first which KNAT1 subsequently lost or it is a function that STM has gained has not been proven although the former appears more plausible as repression of AS1 is vital in the meristem.
The search for a regulator of STM has been inconclusive there has work performed using differing auxin concentrations (Reinhardt et al 2000). Auxin transport was blocked either genetically or via the use of inhibitors in the SAM which failed to produce lateral organs and there was low expression of STM like genes. When auxin was applied directly to the SAM leaf primordia developed at the site of application. Limited epistasis experiments have suggested that CUP-SHAPED-COTYLEDON (CUC) activity promotes STM expression (Aida and Takada et al 2001). stm mutants have partially united cotyledons like those of cuc1 and cuc2, CUC genes might act partly via STM to repress lateral organ development. However a model which promotes localised build-ups of auxin in cells and prevention of neighbours from forming lateral organ primordia by depleting their auxin concentration appears to be the most favoured.
Williams (1998) was one of the first to suggest that another method of KNOX1 gene regulation was through chromatin remodelling. He noted that differences in tissues and organs to respond to a given KNOX1 gene product might be because the chromatin is arranged for different gene expression between two tissues, making the KNOX1 binding sites more or less accessible. Ori et al (2000) further this argument by showing that mutations in the chromatin remodelling factors SERRATE (SE) and PICKLE enhance the as1 phenotype. It is thought that the remodelling factors make the targets of KNAT1 and KNAT2 more accessible in the leaf primordia and contribute to the enhanced phenotype seen. Although there have been many conceptual models for this event there has been no direct evidence that for the regulation of KNOX1 expression via chromatin remodelling.
Model for possible interactions of KNOX1 genes within the plant SAM (Tsiantis 2001)
KNOX1 genes have been implicated in roles other than meristem maintenance and leaf primordia initiation; however there are many inherent problems with finding out the in vivo functions of KNOX1 genes, primarily the pleitotropic effects of modifying a gene from a plants conception. Work done on transgenic tobacco plants transformed with the rice KNOX1 OSH1 gene has shown a general upregulation of photosynthetic genes making it difficult to identify the primary function of the gene in question. What is therefore needed is a methodology in which a gene of interest can be induced at any time during development. Sakamoto (2001) created a chimeric protein under the control of the CaMV 35s promoter between a KNOX gene Nicotina tabaccum homeobox 15 (NTH15) and the human glucocorticoid receptor. The fusion protein was inactive unless in the presence of the glucocorticoid receptor molecule dexamethasone (DEX) where the NTH15 part of the protein was reactivated and therefore expressed ectopically. Results showed that NTH15 bound to and suppressed expression of giberellin biosynthetic gene GA20 oxidase. By using mobility shift assays and mutagenised primers the exact binding sequence of the homeodomain was found. This is clearly the way ahead in term of elucidating functions of KNOX1 genes but it has limited value as it is still not observing the gene in its wild type form, however this could be modified to create a transgenic plant that had its wild type gene removed and replaced with a steroid based inhibitor fusion protein. (However I am not sure about the practical applications of this.)
There have not been many studies on the exact roles KNOX1 genes play other than the above mentioned although there are a few, mainly noted from rescue experiments that did not manage to entirely rescue the phenotype implying secondary roles for some KNOX1 genes. STM is thought to have a role in correct floral development as KNAT1 is unable to form complete flowers in stm as1mutants (Ori et al 2000). PHAN is also thought to play a role in specifying the dorso-ventral axis of leaves although this is not seen in maize and Arabidopsis do not display radial leaves (Timmermans et al 2000).
The most interesting auxiliary use of KNOX1 genes is inferred from the correlation between expression pattern and leaf morphology. Bharathan (2002) looked at KNOX1 gene expression in various vascular plants and the leaf primordia, complex leaf primordia can mature into simple leaves so this was the most informative. All species showed an downregulation of KNOX1 genes at sites of leaf initiation but differed in whether or not KNOX1 genes were expressed during development. In simple leaved species such as maize rice and Arabidopsis KNOX1 genes were expressed in the SAM and downregulated after leaf initiation, over expression of KNOX1 genes resulted in the familiar distortions of the leaf tissue. Complex leave species such as the tomato showed an upregulation of KNOX1 genes in the leaf primordia and a downregulation the leaf. Ectopic expression resulted in more complex leaves. This applies to all vascular plants except a group of legumes in which there is no KNOX1 expression in complex leaf primordia, but it is believed that another gene PEAFLO has taken the place of a KNOX1 gene in the cascade.
The best evidence for this observation comes from a single species N. aquatica (Brassicaceae) an aquatic species with both simple or complex submerged and aerial leaves. Simple leaves occur under high light intensities and show a low level of KNOX1 expression whilst the more complex leaves arise from low light conditions and correlate with a high KNOX1 expression. It is hypothesised that the KNOX1 expression may be controlled by the light which in turn alters plant hormone levels.
Given this data they inferred that a complex leaved based common ancestor had complex primordia with early KNOX1 expression and that simple leaves evolved either through turning off KNOX1 in the primordia or by modifying secondary morphogenesis.
The evolutionary history of the KNOX1 genes has only been touched upon so far although there is little information that is available. Sufficed to say that it is believed that the KNOX1 genes arose though a single common ancestor and have since duplicated. This can be seen in Arabidopsis where KNAT2 and KNAT6 are located within segmental chromosomal duplications. There is also evidence of very high conservation of the homeobox domain, which is not surprising considering they are transcription factors with DNA binding domains.
Discussion
KNOX1 genes have been shown to be instrumental in the maintenance and proliferation of the meristem, through their inhibition pivotal in the formation of a lateral organ (leaf) and also play a role in determining the morphology of the leaves themselves. They exhibit partial redundancy, which while not complete for the plant allows successful elucidation of other roles they play in plant development. There is a lot to be discovered about their regulation and most of what has been shown is based on models but with new techniques such as chimeric proteins and transgenics the answers will not evade us for long.
Bibliography
Bharathan G, Goliber TE, Moore C, Kessler S, Pham T, Sinha NR. Homologies in leaf form inferred from KNOXI gene expression Science. 2002 Jun 7;296(5574):1858-60.
Golz JF, Keck EJ, Hudson A Spontaneous mutations in KNOX genes give rise to a novel floral structure in Antirrhinum. Curr Biol. 2002 Apr 2;12(7):515-22.
Golz JF, Hudson A. Plant Cell 2002;14 Signalling in plant lateral organ development.
Byrne ME, Simorowski J, Martienssen RA. ASYMMETRIC LEAVES1 reveals knox gene redundancy in Arabidopsis. Development. 2002 Apr;129(8):1957-65.
Tsiantis M. Control of shoot cell fate: beyond homeoboxes. Plant Cell. 2001 Apr;13(4):733-8.
Barton M. Leaving the meristem behind: regulation of KNOX genes. Genome Biol. 2001;2(1):
Sakamoto T, Kamiya N, Ueguchi-Tanaka M, Iwahori S, Matsuoka M. KNOX homeodomain protein directly suppresses the expression of a gibberellin biosynthetic gene in the tobacco shoot apical meristem.Genes Dev. 2001 Mar 1;15(5):581-90.
Williams RW.Plant homeobox genes: many functions stem from a common motif. Bioessays. 1998 Apr;20(4):280-2.
Kerstetter RA, Hake S. Shoot Meristem Formation in Vegetative Development. Plant Cell. 1997 Jul;9(7):1001-1010
Long JA, Moan EI, Medford JI, Barton MK. A member of the KNOTTED class of homeodomain proteins encoded by the STM gene of Arabidopsis. Nature. 1996 Jan 4;379(6560):66-9.
