Introduction   
In Hokkaido in recent years, fast-growing willow trees have gained attention as woody biomass resource crops, and in some parts of Hokkaido, surveys have been conducted on initiatives and uses aimed at establishing techniques for their stable cultivation. Until now, willows have been considered for the production of bioethanol as a renewable energy source, for use as a heating fuel, and for livestock bedding. By promoting the multifaceted use of willows, they are expected to be utilised as a regional resource with unused land dedicated to their cultivation. Further, because willows have strong coppice regeneration, are easy to grow and can be managed during short logging periods, they are expected to be a source of broadleaf sawdust for the substrate cultivation of mushrooms such as shiitake where there are concerns about insufficient supply.

Until now, the authors have focused on the willow species Salix udensis and Salix pet-susu, which have the potential for expanded cultivation, and reported that they are useful as base substrates in shiitake cultivation, and that the inclusion of willow bark may have a yield-increasing effect in shiitake cultivation.
The aim in this study was to expand the use of willow sawdust beyond shiitake and target the relatively popular Pleurotus eryngii (king oyster mushrooms). P. eryngii have been actively cultivated since the late 1990s, and their domestic production is third, behind maitake. It is believed that not only increasing productivity in P. eryngii cultivation with high production volumes but also differentiating based on the quality of the fruiting bodies can contribute to improving its competitiveness. Therefore, we aimed to clarify the applicability of willow as a substrate material by evaluating the effect of the type of base substrate on the generation and flavour profile of P. eryngii as an effect of willow sawdust on substrate-cultivated P. eryngii.

1.    Test strain
The P. eryngii strain purchased and tested was EG-109 (made by Kinokkusu) .

Japanese white birch (Betula platyphylla var. japonica. Hereinafter referred to as “birch”), larch (Larix kaempferi), coarse and fine Salix udensis sawdust, coarse and fine Salix pet-susu sawdust and corn cob meal (ground corn cobs, hereinafter referred to as “corn cob”) were used as the base substrates.
The sawdust was made from both Salix udensis and Salix pet-susu willow trees 10-20 years old (5-6m in height), which consitituted the raw materials for the sawdust, with the bark left on the wood. The birch and larch sawdust were made from wood with the bark removed.
The average sawdust particle size was 1.30 mm for the birch, 1.08 mm for the larch, 1.54 mm for the coarse Salix udensis, 1.04 mm for the fine Salix udensis, 1.51 mm for the coarse Salix pet-susu, 1.14 mm for the fine Salix pet-susu, and 2.11 mm for the corn cob. Rice bran and wheat bran were used for the nutrition supplement.

In Test 1, the objective was to assess the influence of base substrate in bag cultivation by comparing the cultivation results of willow with tree species that are easily obtainable and suitable for mushroom cultivation within Hokkaido. In Test 2, the objective was to also include corn cob, which is often used in P. eryngii cultivation, and prepare substrates of varying amounts for use in bottle cultivation and then compare the cultivation results. In Test 3, the influence of the base substrate in bottle cultivation was assessed by comparing the cultivation results with the appropriate amount of substrate in the bottles. The substrate in Tests 1 and 2 was composed of 60% moisture, 24% base substrate, 8% rice bran and 8% wheat bran. In Test 1, birch sawdust was used as the control substrate. The test substrates used larch sawdust and sawdust from four types of willow. The six different substrates were put into polypropylene cultivation bags (Miki Sangyo, Mikipack MP13-C-25, hereinafter referred to as “PP bags”), filled with 1kg of substrate per bag. The bags were then autoclaved at a maximum temperature of 121 degrees Celcius, held for 30 minutes before being allowed to cool. In Test 2, substrate using larch sawdust, corn cob and willow sawdust (a mix of equal amounts of coarse and fine of each tree species) were prepared as the base substrates. To avoid gaps in the bottles, test groups with varying amounts of substrate in each one were set up, with a total of 12 types of substrate prepared: three bottles with 500-540g of larch substrate, three with 525-580g of corn cob substrate and six with 480-520g of willow substrate. In Test 3, the corn cob only substrate was the control group and the substrates using willow sawdust (with a mix of equal amounts of coarse and fine for each tree species) were the test groups. The moisture content of the substrate was 60% and 40g each of rice bran and wheat bran was added to each bottle as the nutrition supplement. Based on the results of Test 2 and the filled state of the substrate in the bottles, the corn cob substrate was set at 540g per bottle and the willow substrates at 500g per bottle. For the bottle cultivation of Tests 2 and 3, a prescribed amount of each medium was put into 850mL cultivation bottles (hereinafter referred to as “PP bottles”), sterilised under high pressure and then left to cool.

The spawn was inoculated then incubated at 22 ± 1 C a relative humidity of 70 ± 10% and under dark conditions for 35 days (for bottle cultivation) and 42 days (for bag cultivation). Once incubation was complete, they were scraped with a flat scraper and the substrate blocks moved to a growing room with a temperature of 16 ± 1C and a relative humidity of 85 ± 10%. In the late growing regulated with 350 lux (using a white fluorescent lamp FL20S-W from Toshiba Lighting and Technology Corporation) for 12 hours a day, the fruiting bodies grew and were then harvested once the top of the caps became flat. At the time of harvest, the appearance of the fruiting bodies was observed and their raw weight measures. The number of samples in each test group was 10 bags (in Test 1) and 16 bottles (in Tests 2 and 3).

Air-dried sawdust (with a particle size of 0.25-0.5mm) was tested. The substrates were analysed for their content of ash, carbon, nitrogen, extracts with diethyl ether and 50% (v/v) ethanol, lignins (acid-soluble and acid-insoluble), carbohydrates (glucan, xylan, mannan, arabinan) and total phenols in the 50% (v/v) ethanol extract solution as previously reported. This was repeated three times.

The data from the cultivation tests were shown as mean ± standard deviation and multiple comparison tests were conducted, Dunnett for Test 1 and Tukey for Tests 2 and 3, using Excel Statistics (by Social Survey Research Information). Further, principal component analysis was used to analyse the data from the component analysis.

Further, as there was no clear difference in the cultivation results between the coarse and fine sawdust in the willow groups, a mix of equal amounts of fine and coarse sawdust was used in the subsequent cultivation tests. Additionally, since the cultivation results in the larch group were the same or better than those in the birch group, the test groups excluding a birch group were set up in the subsequent cultivation tests and compared to the corn cob groups.

Cultivation Test 2 was conducted by filling the PP bottles with each of the 12 types of substrate of different wood species sawdust in various amounts. In this test, the corn cob group was the control group and was compared with the larch and willow groups. No major difference was observed in the appearance of the fruiting bodies (Fig. 1).

In terms of fruiting body yield per bottle, the yields from the willow groups were 0.89-1.09 times that of the corn cob group (525g/bottle) and 0.83-0.93 in the larch groups. Converted to yield per 1kg of substrate, the yield from the willow groups was 0.92-1.14 times that of the corn cob group (525g/bottle) and 0.84-0.91 times in the larch group (Table 2). The yield in the willow groups was similar to that of the corn cob group, but with less substrate, and the Salix pet-susu group (500g/bottle) had a greater yield than the larch groups (500g/bottle, 520g/bottle) (with a statistical difference). However, the yield varies depending on the amount of substrate and there is an optimal amount.

Based on the results from Test 2, an optimal amount of substrate was established, three different types of base substrate were put into PP bottles and cultivation Test 3 was performed. In terms of fruiting body yield per bottle, the Salix udensis yield was 0.97 times that of the control group and the Salix pet-susu was similar at 0.94 times. Converted to yield per 1kg of substrate, the yield from the Salix udensis group was 1.04 times that of the control group and the yield from the Salix pet-susu was 1.02 times (Table 3). The willow groups had similar cultivation results to the control group, which used corn cob. In P. eryngii cultivation, the effectiveness of using wholewheat bran in place of wheat bran has been reported, showing the clear influence the type of nutrition supplement used in the substrate has on yield.

The results of the base substrate analysis are shown in Table 4. The content of each component is arranged from the lowest amounts, found in the larch, to the highest, found in the corn cob, and highest to lowest for carbon, lignin and glucan, with the values for the two types of willow in the middle. The results of this analysis, as analysed by principal component analysis, are shown in Figures 2 and 3. From the principal component loading in the principal component analysis (Figure 2), the first principal component had a strong positive correlation with lignin, carbon and glucan, and a strong negative correlation with extracts, ash, nitrogen and xylan. The second principal component had a strong positive correlation with total phenols.

From the main component score of each raw material (Figure 3), the corn cob point is plotted at a distance from the other sawdust points, while the Salix udensis and Salix pet-susu are plotted close to the origin, showing that they occupy an intermediate position among the raw materials.

The two types of willow have a higher nitrogen content compared to the other tree species, although not as high corn cob. Further, the fine sawdust of the two willow species resulted in a higher nitrogen content than the coarse sawdust, but this is presumed to be due to the higher proportion of bark, which has a high nitrogen content, mixed into fine sawdust. In P. eryngii cultivation, it has been reported that there is an increased yield effect when the lees from shochu distillation is used as a nutrition supplement and the substrate has a high nitrogen content. Further, favourable cultivation results have also been achieved in golden oyster mushroom cultivation using the Sasa veitchii bamboo, which has a high nitrogen content but a low C/N. Also, when spent (shiitake) cultivation logs with a high nitrogen content but low C/N were used at the base substrate for oyster mushroom cultivation, the result was an approximately 20% increase in annual yield (yield/bottle/year), a measure of productivity, compared to sawdust-based substrate. In addition, in shiitake substrate block cultivation where sawdust without bark is replaced with willow bark within a certain range, the results suggest a relationship between the nitrogen content of the substrate and an increase in yield. The relatively high nitrogen content of willow sawdust is thought to be one of the reasons why the cultivation results are similar when using willow sawdust for P. eryngii cultivation and when using corn cob.

Until now, it has been reported in shiitake and bunashimeji that the taste of cultivated mushrooms is strongly influenced by the mushroom variety. On the other hand, it has been reported that in P. eryngii cultivation fruiting bodies grown in substrate supplemented with shochu lees have more umami and are sweeter compared to those cultivated in conventional substrate hence they are strongly influenced by the nutrition supplement, and there is a higher amino acid content in the fruiting bodies. This suggests that the high nitrogen content in nutrition supplement may influence the flavour profile.

Further, the results of evaluating the effect of substrate amount in bottle cultivation showed that the yield (per unit weight) in the willow groups was up to 1.14 times that in the corn cob group and that a similar yield was produced with less substrate than the corn cob group. From the component analysis of the base substrates provided for the cultivation tests, the characteristics of the component composition of the two types of willow tree were clearly different from the other materials, showing that they were positioned in between the other materials. The results of the sensory evaluations of P. eryngii from the cultivation tests showed that the P. eryngii from the willow groups were highly rated for taste and texture. Based on the above results, Salix udensis and Salix pet-susu sawdust are applicable in the substrate cultivation of P. eryngii.