Hagberg falling number and breadmaking quality

ABSTRACT

This review includes major contributions from the Flour Milling and Baking Research Association (FMBRA), the National Institute of Agricultural Botany (NIAB), the Institute of Plant Science Research (IPSR) (Cambridge Laboratory) and the Edinburgh School of Agriculture, in addition to a section contributed by colleagues in the Agricultural Development and Advisory Service (ADAS).

The EEC policy of encouraging home production of quality wheat has resulted in an increase in the price differential between home-produced and imported wheat so that millers are seeking to use the maximum proportion of home-produced grain in their grists. Some deficiencies in the home-produced grain have been overcome by supplements of gluten or modification of the baking process. However a major restraint to any further increase in the use of home-produced grain remains the high level of alpha-amylase enzyme present in many samples, particularly in cold and damp summers. The Hagberg Falling Number (HFN) test is used as an indicator of the level of alpha-amylase enzyme in the sample.

MILLING AND BAKING
The enzyme alpha-amylase has long been known to be produced in the early stages of germination. It is the most important of a group of enzymes that break down complex food reserves in the grain into simple soluble sugars needed by the developing embryo. At the onset of germination the level of alpha-amylase enzyme can undergo a thousand-fold increase and the presence of only a few such grains can affect the HFN of a large bulk of grain. High levels of alpha-amylase enzyme result in loaves with a sticky interior which causes problems for commercial bakers' slicing machinery.

The HFN test has been developed to provide a reasonably rapid and cheap indication of the level of alpha-amylase enzyme in the grain. However, to provide reproducible values it involves very precise milling of the grain, correction of sample weight for moisture content and strict compliance with procedures for the use of test equipment. Since only 7g of flour are used in the test it is also imperative that the sample accurately represents the bulk from which it is drawn. The test is accurate to +- 5% if conducted correctly, although greater errors are frequently observed because of the difficulties in obtaining representative samples and insufficiently rigid control of the test. It is also important to recognise that HFN numbers are not directly additive. The result of mixing two samples can be estimated but the mixture will have a much lower HFN than a direct mathematical mean.

FMBRA has investigated many methods of minimising the deleterious effects of low HFN but the benefits are generally marginal except in the case of simultaneous combination of microwave and thermal baking. Unfortunately, this method has not as yet been commercially acceptable mainly because of cost.

VARIETIES
HFN is an important factor in wheat breeding programmes and is one of the quality tests that is applied to all varieties within the UK National List and Recommended List testing schemes. After the initial screening year, varieties are grown with and without fungicides. In the years 1981-83 HFN tests were conducted on samples from both treatments but, since then, only the fungicide-treated wheats have been tested. It has long been recognised that pre-harvest sprouting has profoundly deleterious effects on HFN so, since the early 1950s, varieties have been assessed for predisposition to pre-harvest sprouting.

It is apparent that among current varieties, spring cultivars tend to have better HFN values but this is thought to be due to the higher standards that have been required for the minority spring crop to find acceptance. Variation from season to season and site to site has been extremely large in both winter and spring varieties.

While variation from site to site or season to season often had little effect on the ranking order of varieties, it was observed that some varieties were more variable than others. In some cases a low resistance to sprouting is thought to contribute to variable HFN values but in other cases a clear relationship is not established suggesting that in some conditions alpha-amylase is produced independent of visible sprouting.

THE INFLUENCE OF HUSBANDRY
During the last few seasons the HFN test has been included with other routine tests of samples from a wide range of ADAS husbandry trials. In many instances a relationship would appear to exist between treatment and HFN but further detailed examination points to only an indirect effect. For example, at first sight it would appear that growth regulators have improved HFN values but this is actually an indirect result of controlling lodging. Similarly, cultivations can have an effect by influencing the survival of volunteers from a previous crop. Time of sowing trials need to be harvested promptly to avoid the early sowing being at risk to adverse weather. When such details were examined, there was no reliable direct relationship with cultivation, date of drilling or growth regulators.

Fungicides have an effect which seems to be greatest if they are used repeatedly and achieve a marked increase in duration of green leaf area. Moreover, there is evidence that some groups of fungicides have more effect on HFN than others, suggesting a direct growth regulatory effect on enzyme production.

Analysis of agronomy trials has shown that if spring nitrogen application is below the optimum for yield, the HFN will also suffer. Often the response of HFN to nitrogen appears to be dependent on varieties.

Other work has concentrated on predicting HFN values at harvest by extrapolation from samples taken 10-14 days earlier when the moisture content is still 25-30%. Such a prediction service, when perfected would allow the crops from fields with low HFN to be stored separately. The better fields could be given priority and their quality preserved.

A detailed study has been made of the effect of one commercial gravity separator and it was shown to be capable of separating a mixed sample very effectively so that the HFN of a proportion (often 60-70%) of a sample starting with a value of less than 100 can be raised to over 200 while the remainder still sells at a feed wheat price. However, it has also been noted that the breadmaking value of this separated wheat may not be improved to the extent expected from the improved HFN values.

GENETICS AND BREEDING
In recent years, the seasons of 1977, 1985 and 1987 were noted for the high levels of alpha-amylase (low HFN) recorded at harvest. It has been shown that enzyme production has been the result of different causes in each of these three seasons. In 1977 there was conventional sprouting of mature grains post-dormancy but, in 1985, the enzyme production was much earlier in the development of the grain. In 1987 the problem was germination following the onset of ripening but prior to the onset of dormancy. Furthermore, there is evidence that the three different physiological mechanisms are each under independent genetic control.

The mechanism of inheritance in grain tissues, together with the difficulties of manipulating genes to alter levels of alpha-amylase in sprouted grain is discussed. It is likely that the several key UK varieties which have inherited a susceptibility to the expression of pre-maturity alpha-amylase enzyme obtained the genes responsible from the Belgian variety Professeur Marchal. It is suggested that the genes responsible for high pre-maturity alpha-amylase can be removed by judicious selection.

The gibberellin-insensitive dwarfing genes including Rhtl, Rht2 and Rht3 all have some action in reducing pre-maturity alpha-amylase levels although only the more extreme Rht3 dwarfing gene is likely to have an economically significant effect agricultural practice.

The difficulties of measuring dormancy in field conditions are discussed and it is concluded that red grain colour is associated with increased dormancy because either the red phlobaphene pigment itself, or precursors of it, cause a temporary inhibition of germination. Selection for red colour as opposed to white is obviously simple. However, it is likely that 3-gene reds would have greater dormancy than single gene reds and the distinction between 1, 2 and 3-gene red wheats is more difficult. Further sources of dormancy are discussed but it is not yet clear if these can all provide cumulative benefit.

The relationship between pre- and post-maturity sprouting is not understood, thus it is not known whether the same genetic strategies to achieve resistance should be employed.

BIOCHEMISTRY
As previously indicated, there are at least three periods when alpha- amylase enzyme is produced in response to the appropriate stimulus but the appropriate combinations do not occur regularly and therefore the mechanisms have not been well researched. However, it is suggested that variation between varieties may be the result of numerous mechanisms. These include physical differences in the grain surface which affect water and oxygen absorption, and differences in the rate of embryo development. Many complex biochemical processes are occurring in the developing grain and inevitably these will affect the enzyme production systems. Probably the most investigated but still not completely understood group of effects are those associated with endogenous plant growth hormones, including gibberellins, auxin, cytokinins and ABA. The current understanding of these growth hormones and the ways in which their action is modified by environmental conditions is reviewed.

FUTURE RESEARCH AND DEVELOPMENT REQUIREMENTS
Detailed recommendations are made under each section but include the following:-

While HFN has been used as a convenient rapid indicator of enzyme levels and suitability for breadmaking, there are indications that other factors can be influential. For example, HFN can in some situations be improved by gravity separation while breadmaking may not be improved. This needs further investigation.

It is essential that a well co-ordinated national system of variety testing is maintained with samples from a wide range of environments being tested to evaluate the potential level of HFN and stability over a wide range of conditions.

Having identified differences in susceptibility to pre-germination enzyme production, the biochemistry of this system should be thoroughly researched and the benefits transferred to plant breeding as soon as possible.

Since the HFN test does not always reflect precisely the value of wheat for breadmaking, alternatives should be investigated, preferably identifying a test that could be used by breeders and variety testing authorities as well as grain purchasers.

HFN results from stored grain have been rather inconsistent and a more thorough investigation is necessary to determine the effect of storage conditions and moisture content.

The effect of fungicide applications in lowering HFN values needs further investigation to determine optimum strategies which could vary for different varieties, diseases and environments. The use of a desiccant should be included as a treatment in appropriate fungicide, nitrogen and time of harvest trials so that its value in aiding the final drying of the crop is more thoroughly investigated.

If a reliable pre-harvest sampling service is to be available further work is necessary to confirm details of the appropriate sampling technique.

Further research is required to determine precisely the environmental factors that trigger pre-maturity alpha-amylase production so that unequivocal screening methods can be made available for use in research and plant breeding. The major genes involved in the enzyme system should be characterised and the need for still further genetic protection evaluated.

Further work is necessary to determine the extent of the dormancy regulating potential of the R genes that provide red seed coat colour. A better understanding of their mechanism will provide guidance for their most effective use.

Pre-maturity sprouting in wheat is a rare phenomenon which caused problems in 1987. Further work is required to identify precisely the environmental trigger involved before real genetic analysis can proceed.